Luca Dellanna's Blog http://localhost:8080 People management and risk management. Wed, 19 Nov 2025 10:52:29 GMT https://validator.w3.org/feed/docs/rss2.html https://github.com/jpmonette/feed en All rights reserved 2025 <![CDATA[A one-minute exercise to de-risk your life]]> http://localhost:8080/posts/derisk-exercise http://localhost:8080/posts/derisk-exercise Mon, 13 Nov 2023 00:00:00 GMT <![CDATA[Culture is the track record]]> http://localhost:8080/posts/culture-is-the-track-record http://localhost:8080/posts/culture-is-the-track-record Sat, 23 Sep 2023 00:00:00 GMT Leaders who think organizational culture is a set of concepts attempt to change it using words and concepts – and inevitably fail, because organizational culture is not a set of concepts. It's a track record. In particular, it's the track record of which behaviors are a waste of time and which lead to good personal outcomes. For example, what happens when someone raises a hand – do they get listened to, or do they learn that raising a hand is a wasted effort? In other words, what’s the track record of what happens when someone raises a hand? That’s what determines whether raising hands is part of the team’s culture. Defining Core Values and communicating them doesn’t change cultures. It’s not a useless exercise but merely a first step. After Core Values are defined, track records must be changed. This cannot be achieved through words, only actions. Or, more precisely, words can initiate and facilitate the process, but only words backed by action count. ### Engagement is also a track record It is the track record of what happens when one cares. What happens when someone works hard? When someone points out a problem? When someone comes up with an idea? Does it lead to good or bad outcomes? What's the track record of caring? Does it lead to good personal outcomes, or does it lead to wasted effort? If you want your people to be more engaged, change the track record of caring. Make sure that the next time someone cares, good things happen to them. Or at least they aren’t taught the lesson that it would have been better to care less. ### Teamwork is a track record Similarly, and contrary to common belief, teamwork is not about liking or trusting your colleagues. Instead, teamwork is the track record of what happens when colleagues interact. What happens when someone asks a colleague for help? Does what follows teach them that it was a good idea to ask for help, or that asking for help is a waste of time? What happens when someone gives feedback to a colleague? Are they listened to and thanked? Or are they made wish that they hadn’t voiced their feedback? Again, to improve teamwork, improve the track record of interactions. ### Improving the track record Let’s work on this last point. How do you improve the track record of interactions? The trick is to not address all interactions at once – such a generic goal will produce a generic approach that won’t be effective. Instead, begin by picking one type of interaction and working on that. For example, let’s work with the interaction of “asking and receiving feedback.” People won’t ask for feedback unless, in your team, there is a track record that, when people ask for feedback, they receive helpful and actionable feedback that doesn’t feel personal. And people won’t give good feedback unless the track record in your team is that, when people give feedback, it is well received and taken seriously. So, if you want your people to give more feedback to each other, you need to create these two track records. ### Leading by example Achieving this requires, first and foremost, that your personal actions contribute to the new track record. Whenever you give feedback, make sure it’s not just correct but also helpful, and that makes the receiver glad to have you as a manager rather than wishing you didn’t exist. And whenever you receive feedback, make sure you take it seriously. This doesn’t mean accepting all feedback as correct – some will be wrong – but always making your interlocutor feel listened to, and if you disagree, let them know why. Never take any feedback you receive personally, and never make any colleague giving you feedback feel like they wasted their time. The more your actions show that in your team there is a good track record associated with giving feedback, the more people will give and request feedback. ### Teach skills However, leading by example is necessary but not sufficient. Not only must your people be open to giving and receiving feedback, but they must also have the skills to do it in a helpful way that makes their interlocutor want to have more such interactions in the future. This requires you to train and coach them on how to give and receive feedback. "Politically correct" actions, such as forcing people to say thank you even when they receive bad feedback, won't work. Instead, what will work is to teach your people to give such helpful feedback that saying "thank you" is a natural reaction. ### Conclusions So, to summarize what we’ve seen so far. Organizational culture – what your team does, what your team cares about, what your team doesn’t do – is not a bunch of words but a track record of actions and reactions. So, changing your organizational culture means changing the track record. Find out the behaviors you want your team to exhibit, and ensure that there’s a track record of good things happening to those who exhibit them. ]]> <![CDATA[The Depolarization Manifesto]]> http://localhost:8080/posts/depolarization http://localhost:8080/posts/depolarization Sat, 26 Apr 2025 00:00:00 GMT Voters lose their focus when they believe that winning the next election is more important than electing a competent government. This occurs when voters view a particular issue as a life-or-death situation and allow this perception to lower their standards for what they expect from representatives, including their own. In their urgency to ensure their party wins, they tolerate incompetent and corrupt candidates, overlook their mistakes, support their flawed policies, and abandon the fight for truth and competence. If this occurred for just one election, the damage might be limited. However, when every election is framed as a matter of life or death, and voters progressively accept worse and worse behavior, the quality of government rapidly deteriorates. ### III The problem with government quality deteriorating is not only that citizens become poorer and receive worse public services, but also that it increases the likelihood of dissatisfied voters switching sides, even if the opposition runs terrible candidates. Compromising on standards to help your party succeed might win the next election, but it ultimately leads to losing future ones, resulting in incompetent and polarized political opponents in power. A disaster. This is why it is never worthwhile to lower your standards for your party’s candidates, even when preventing the opposition from winning feels like a matter of life and death. ## Rationale ### I Polarization is not just political differences or heated debate. It’s the belief that the opposing group poses such a danger that it justifies adopting unethical behavior against them: insulting, lying, cheating, suspending due process, collective punishment, and so on. ### II Polarization is fear-driven but not irrational. Its logic is that the other group is so dangerous that extreme measures are justified to limit their influence and negative impact. However, adopting extreme measures makes one’s own group appear unreasonable and dangerous, thereby justifying extreme measures from the opposition. Therefore, polarization is always mutual. ### III If polarization weren’t mutual, it would be an effective tool for advancing political goals. However, since polarized behavior from one group triggers similar responses from others, it gradually erodes societal trust and institutional effectiveness. As a result, governing becomes more difficult even if one wins the elections, and society suffers even if one gets to pass their favorite policies. For example, if someone sees environmental neglect as a life-or-death risk, wanting to use any method to raise awareness is comprehensible. However, if these methods alienate most of the population, it becomes increasingly difficult, if not impossible, to gain the broader societal support required to improve the environment at scale. Therefore, it is paramount never to let virtuous ends justify polarized means. ### IV Under a short enough time horizon, using polarizing tactics always makes sense, and under a long enough time horizon, it never does. For example, using censorship against a wannabe dictator is a good idea to prevent his rise to power. The question is whether normalizing censorship makes it easier for another wannabe dictator to rise to power. So, of course, there are good arguments for why the use of polarizing tactics brings good short-term outcomes. The question is whether the long-term effect is positive, and the answer is inevitably negative. ## Escalation ### I Polarization begins when someone expresses a belief that’s perceived as dangerous by the outgroup, and their ingroup fails to dissociate from them. For example, when someone makes a racist comment, and their community members do not call them out on it. This silence makes the entire ingroup appear dangerous to the outgroup. ### II The perceived dangerousness of a group stems from two factors: (1) the group harbors unethical individuals, and (2) the group seems unwilling or unable to call them out or hold them accountable. A third factor often exacerbates this: when the group conspicuously celebrates their unity, including unity with those unethical individuals. ### III There are several reasons the ingroup may be unwilling to police its members who do or say unethical things: They justify it as “tit for tat.” They view the person making such statements as “that weird guy” and cannot imagine others would take them seriously or consider them representative of the whole group's beliefs. They are already so polarized that they believe they face a life-or-death situation where it is paramount to stick together and offer no vulnerability to their political opponents. Whether these three behaviors are justified is irrelevant. What matters is whether they invite similar behavior from the other group. ### IV And yet, the very people who believe (1), (2), and (3) above often fail to extend the same benefit of the doubt to the outgroup. When someone from the outgroup says something polarizing, they take it at face value. When “that weird guy” from the other group says something dangerous, they present it as if coming from a respected representative who speaks for the entire group. ### V These behaviors intensify polarization and make all groups more likely to engage in further polarizing actions and polarizing rhetoric. A race to the bottom where even the winners lose. ## Temptations ### I Because polarization is always mutual, it cannot be resolved through polarizing actions. For instance, censoring a political opponent only deepens polarization, as a portion of the population will view the censorship as unjustified and, consequently, perceive the censors themselves as the real threat. ### II It is paramount to resist the temptation to say, “The other side exaggerates and misrepresents, so we are justified in doing the same” – as the saying goes, an eye for an eye, and the world goes blind. ### III It is paramount to resist the temptation to call for censorship. Not only is focusing on what people say rarely the best approach to address hate speech (enforcing existing laws against criminal actions would put most hateful individuals behind bars anyway, with far less collateral damage to free speech), but censorship itself is polarizing and invites retaliatory censorship against you. ### IV It is paramount to resist the temptation to claim, “The other side is so dangerous, we must stop them at all costs.” Is the entire opposing side truly dangerous, or only certain members? And if only some are dangerous, is exaggeration really the most effective way to get their side to distance themselves from these individuals? Accurately representing the dangers posed by specific individuals often makes a more compelling argument that can garner broader support than exaggeration ever could. ### V It is paramount to resist the temptation to say, “If we do not exaggerate and misrepresent, we will lose the elections.” If you resort to exaggeration and misrepresentation, you will get pigs elected, regardless of the color of their tie. Not only will electing a pig make you miserable even if they share your party colors, but it will increase the chances of your party losing subsequent elections, as dissatisfied voters switch to the other side. ## Depolarization ### I The first step to reduce polarization is to recognize that one's political opponents comprise two distinct groups, reasonable individuals and unreasonable ones, and to avoid conflating the two. Then, limit unnecessary engagement with the unreasonable ones, especially on social media, as interacting with these individuals is precisely what might tempt you to say or do polarizing things. And when you do engage with the unreasonable, always treat them as individuals and never do or say things implying you believe everyone in their community is unreasonable. ### II The second step to reduce polarization is to always address specific individuals when criticizing behavior from the opposing group, and never misrepresent or exaggerate their actions. Any generalization will inevitably be perceived as polarizing. ### III The third step to reduce polarization is to always address the reasonable members of the opposing group when proposing policies, explicitly listening to their concerns and addressing them without dismissing them. This approach not only decreases polarization but also helps craft better and more widely acceptable policies that are more likely to be implemented and sustained. This stands in direct contrast to what most of us do when proposing policies: addressing our own group, who already supports them and needs no further convincing – and in the meantime, antagonizing the opposing group, whose support we need for the policy to be effectively implemented and resist the next election cycle. ### IV The fourth step to reduce polarization is to focus on the strongest argument for or against a policy rather than exploring the second-strongest, third-strongest, and weaker points. The further down the list of arguments one goes, the less solid they become, and the more public discussion will fixate on the most divisive details. ### V The fifth step to reduce polarization is to consistently call out unreasonable behavior from one’s own side. Failing to do so not only encourages more such behavior, but can be interpreted by the opposing group as tacit approval – and nothing seems more threatening than an unreasonable individual gathering widespread support. ### VI The sixth step to reduce polarization is to consistently call out exaggerations and misrepresentations, both from people belonging to the other side and our own – in particular from our own, for not calling out our unreasonable allies is the easiest way to make our whole group look dangerous. ### VII The seventh and final step to reduce polarization is to consistently emphasize that polarization is always mutual. That terrible president? They rose to power because of mutual polarization. That lack of trust in institutions? It occurred because of mutual polarization. Even when one side bears more responsibility than the other, we must all address polarization as a shared problem. Failing to do so is not only polarizing in itself, but will lead even the less culpable group to justify unethical behavior. ### VIII The more the term “mutual polarization” enters public discourse, the more we will all realize what is going on, and the more we will avoid using solutions that are themselves polarizing. ## Hypocrisy ### I Reducing polarization to zero is clearly impossible. This doesn’t mean it cannot be significantly diminished. ### II Even the most dedicated adherents of this manifesto will occasionally say or do something polarizing. Critics will inevitably point this out: “You ask us not to polarize, but you did it yourself that time. You’re a hypocrite.” But hypocrisy is overrated. If only the perfect were allowed to call out others, then no one could call out anyone, and the world would quickly descend into mediocrity, chaos, and impunity. The only path toward mutual improvement is to allow the imperfect to call out others. ### III Polarization is a coordination problem. When we encounter something polarizing, we feel compelled to respond with polarizing actions. But if we all stopped engaging in polarizing behavior simultaneously, we wouldn’t feel this compulsion. Committing to a goal even when others are out of sync is the only way to initiate coordination. ]]> <![CDATA[Adversarial vs Collaborative Feedback]]> http://localhost:8080/posts/adversarial-vs-collaborative-feedback http://localhost:8080/posts/adversarial-vs-collaborative-feedback Thu, 10 Apr 2025 00:00:00 GMT The trick is to avoid framing feedback as "me vs you" and instead frame it as "me and you vs a common problem." Let’s see a trivial example. “Your presentation should include more charts” is inherently adversarial, for it implies the recipient was wrong and the manager right. Conversely, “This presentation is very important, and we cannot afford the audience not being as engaged as possible. What if we added some charts?” frames both the manager and the employee as being in a vulnerable situation that requires change – which is proposed as an improvement that does not imply fault. Moreover, the suggestion is not framed as “I arbitrarily decided this is not enough” but rather as “external circumstances require this to be better.” Of course, it’s crucial that none of this is a lie or exaggeration and that the manager genuinely believes in what they are saying. But this is relatively easier, for in most cases, both the manager and the employee are truly fighting the same third-party problem (a competitor, a lack of funds, etc., or even whatever adversary the company’s mission is fighting, such as “lack of access to the company’s product”). Let’s see a second example. “You need to dress better when going to clients.” Again, this implies that the employee is wrong. Moreover, it can be perceived that the manager is unreasonably picky about the dress code. Instead, consider the alternative: “New clients do not trust us enough – and how could they, if they never heard of our products? So, it is paramount we gather all sources of trust, including a professional dress code.” Again, this clarifies the objective need for change, presents it as an improvement without implying the previous approach was necessarily wrong, and positions the manager not as an Ivory Tower almighty sage but rather as a colleague fighting in the same trenches. The less the manager's ego appears in the suggestion, the more the subordinate's ego can accept it. ]]> <![CDATA[The Planner and the Gatekeeper]]> http://localhost:8080/posts/planner-and-gatekeeper http://localhost:8080/posts/planner-and-gatekeeper Sun, 01 Dec 2024 00:00:00 GMT Our brain contains not one but two decision-makers: the Planner (our cortex, which suggests what we should do) and the Gatekeeper (our basal ganglia, which decides whether to do it). The interaction between the two, depicted below, determines most of human behavior. ![The interaction between the Planner and the Gatekeeper](/figures/planner-and-gatekeeper/circuit-open.png) ### A common example Imagine our cortex (the Planner) decides we should hit the gym. If our basal ganglia (the Gatekeeper) doesn’t like exercising, the cortex’s order doesn’t reach our motor areas, and we do not go to the gym. ![The result of liking an outcome but not the action that achieves it](/figures/planner-and-gatekeeper/circuit-closed-with-example.png) What’s the result of liking an outcome but not the action that achieves it? Inaction, and thus, frustration. ### The circuitry The image below depicts a simplified schema of the actual circuitry that you would observe if you dissected a human brain. There is a pathway running from the cortex to the motor areas of our brain. Alongside it, there is an area of our brain called the basal ganglia that can inhibit the pathway, preventing the cortex’s orders from becoming actions. ![The separation of roles](/figures/planner-and-gatekeeper/separation-of-roles.png) This circuitry highlights a separation of roles. The Planner (our cortex) makes plans but does not have direct access to our muscles; therefore, its plans are only suggestions. Conversely, the Gatekeeper (our basal ganglia) cannot make plans but can block the Planner's suggestions from reaching our motor areas. For action to occur, our cortex must suggest it, and our basal ganglia must approve it. If either is missing, we do not take that action. ### Habit formation To create new habits, we must ensure that our cortex sends the desired orders and that our basal ganglia allow them to reach our motor areas. Sadly, most habit formation advice addresses only the former. It involves either making better plans or removing cues from our environment so that we do not come up with bad ideas (such as eating that bag of chips on the table). However, that is cortex-centric advice that ignores the Gatekeeper. No wonder it is not very effective. ### Influencing the Gatekeeper So, what causes the Gatekeeper to open or close the gate? It's simple: experiential memory. The gate opens when the suggested action is remembered to have brought positive emotions in the past. We have already seen that the Gatekeeper cannot imagine or plan, only remember. The part of you that can imagine, plan, and consider indirect or long-term consequences is the Planner, not the Gatekeeper. Knowledge, imagination, and planning help generate good suggestions but do not guarantee execution, nor do they prevent bad ideas created by instincts. Hence, we have two ways to influence the Gatekeeper. First, we can try to feel emotions. If we feel a strong positive emotion when we think about doing an action, we will be more likely to do it. However, note that emotions are only associated with doing the action count, not emotions associated with the reward. That’s why it’s so hard for many people to go for a workout, even though almost everyone has positive associations with looking fitter. Only how we feel about the action counts, not the outcome. (More precisely, how we feel about the outcome only matters to the extent it creates feelings about the action.) The second way in which we can influence the Gatekeeper is by creating new experiences that associate a positive emotion with the suggested action. ### The key principle For us to take action, our cortex must both think about it, and our basal ganglia must feel like doing it would feel good. Hence, habit formation must speak to BOTH the Planner and the Gatekeeper: - To the Planner: offering better knowledge, more cues for positive actions, fewer for negative ones. - To the Gatekeeper: providing new experiences and creating new emotional associations. We must acknowledge that our Planner (our thinking self) cannot just "convince" the Gatekeeper to change. It can only do so indirectly – by planning actions that the GK is already willing to take and having those actions create new emotional associations. **This post is a brief summary of the behavioral model I describe in one of the chapters of my book, "[The Control Heuristic](/books/the-control-heuristic)." Read the book to learn more about it and behavioral change.** ]]> <![CDATA[Reproducible Success Strategies]]> http://localhost:8080/posts/reproducible-success-strategies http://localhost:8080/posts/reproducible-success-strategies Sun, 01 Dec 2024 00:00:00 GMT The key to success is to spend more time on important-yet-not-urgent activities than most of your peers. ### Good long-term strategies are Inevitable Many long-term strategies are variations of the following: undertake a few projects that might fail, but do them in such a way that failure doesn’t compromise future projects, and in such a way that even in failure, you still build know-how, trust, and relationships. If you do so, eventually, you will succeed. As you can see, good long-term strategies often include projects that might fail, but only within the context of a strategy that cannot fail. Do not take “cannot fail” too literally. Whether a 100% success rate is possible is irrelevant. What matters is not to be satisfied with less than that and to continuously ask yourself, “What might cause me to fail, and what can I do about it?” Asking that question might reveal that you are taking too many risks, or perhaps not enough. It might reveal that you are going too fast, skipping steps, or perhaps that you are being too slow or complacent. It might reveal that there is a lesson you don’t want to learn. It might reveal that there is some help you refuse to ask for. It might reveal that you’re using the wrong strategy. From that one question, you can learn a lot. The point is, never treat your strategy as you would treat a tactic. You can tolerate tactics that might not work, but you cannot tolerate strategies that might fail. After all, you only get one life. Make your strategy inevitable. ### Bad long-term strategies We have identified three properties of good long-term strategies: they are sustainable, constructive, and make success inevitable. Now, let’s look at a few red flags indicating a poorly designed long-term strategy: - It is not sustainable: it may provide some early growth but cannot sustain it over time. - It is not constructive: it does not build the long-term assets required for long-term success. - It is not inevitable: it might lead you to success, but it might also not. - It is not actionable: it lacks clear next steps you can work on today. - It relies on the future remaining unchanged: it does not incorporate learning and adaptation. - It focuses on a single area of your life while neglecting the rest. This might lead to a pyrrhic victory, where you achieve your objective, but it may still be insignificant because you have compromised another important area of your life. There are more red flags for a bad long-term strategy, but the ones listed above are the most important. Adding more to the list would dilute its impact and make it less effective. _Note: this was an excerpt from my book, "[Winning Long-Term Games](/books/winning-long-term-games)."_ ]]> <![CDATA[The Dellanna Method]]> http://localhost:8080/posts/dellanna-method http://localhost:8080/posts/dellanna-method Mon, 02 Sep 2024 00:00:00 GMT In most trainings, the trainer talks about some procedure or framework, and the attendees nod, but then go back to their workstations without anything having changed. That's because the training didn't address the real problems of the audience, their bottlenecks to action. Most trainers circle around the hard questions that the audience needs answered: what do you do when there’s not enough time to do everything? What do you do when there’s a subordinate who is fully disengaged? How do you have hard conversations? How do you do it when turnover is already a problem? Instead, my approach is all about surfacing and addressing these hard questions. It’s about acknowledging that the real world is messy and time is a major constraint. It’s about empowering managers by giving them the tools they need rather than the tools we think they need. Moreover, I believe that it's necessary to teach not only what the right things to do are but also how to do them right. Hence the focus on roleplay, on working on concrete examples, and on examining difficult and messy situations. It’s an approach that produces results in the real world. ### Case study #1 I recently delivered a couple of training sessions to the Finance Leadership Team of a major pharmaceutical company. Not only was the engagement during the session excellent (finally, a trainer that addressed their problems), but the growth in people management capabilities has also been evident. ### Case study #2 For nine years and counting, I have been teaching a risk management for manufacturing operations module at an Italian university. I deliver 90% of it using hypotheticals. We go for hours with me asking questions (“what would you do on your first day…?” “what would you do to achieve X?”) and giving motivated feedback (“good idea, because…” or “bad idea, because…”). We also do plenty of roleplay, because one thing is to know at a high level what should be said, and another thing is to be able to say it in a way that is clear and convincing. It works wonderfully. It is more active learning, covers more concrete situations, teaches know-how rather than know-what, and is remembered better than just going through slide decks. ### Example engagement A typical engagement would proceed as follows. - A first call with you, during which I gather information about your team and its day-to-day operations. Ideally, this would also include a couple of calls with members of your team to understand their challenges. - Two 90-minute workshops with your team, where we work through hypotheticals. This is where your people learn how to properly manage common people management situations. - A 30-minute one-on-one session with you and each of your team members. Such an engagement for a team of 8 people would cost approximately 3000€, assuming remote delivery. I would personally conduct all the aforementioned activities, ensuring no bait-and-switch with less experienced consultants. All the activities would be tailored to the specific situations your team faces in their day-to-day work, without using pre-packaged content that is irrelevant to your team’s reality. ]]> <![CDATA[The Dellanna Diagrams]]> http://localhost:8080/posts/dellanna-diagrams http://localhost:8080/posts/dellanna-diagrams Mon, 24 Jun 2024 00:00:00 GMT This is very important because if you merely aim to make yourself or your organization robust to problems, you will promote behaviors that leverage rigidity – a property that decreases antifragility. In fact, a rigid person or organization wants to minimize and avoid problems, whereas antifragile people and organizations do the opposite: they proactively surface problems in order to adapt to them. ### Creating engagement It might seem like the Dellanna Diagram of disengaged students and disengaged employees looks like the one below: problems are either too easy, triggering complacency (the yellow area), or too hard, triggering paralysis or frustration (the red area). ![The Dellanna Diagram – The Fragile](/figures/dellanna-diagram-the-fragile.png) However, their actual diagram is like the one below. There is a green area. It’s just so small that it looks like it’s not there. ![The Dellanna Diagram – The weak antifragile](/figures/dellanna-diagram-the-weak-antifragile.png) This is important because it means that there is a size of problems that is small enough to get acted upon yet significant enough so that solving it triggers learning and increased motivation. Great teachers and great managers excel at finding the right problem to give to their disengaged students and employees, hitting them in the green area and triggering antifragility, progressively growing their skills and motivation until they become skilled and engaged. If you are not convinced it’s possible, think about a 50-year-old person living a sedentary lifestyle. If they suddenly start going to the gym, they might believe that their body’s Dellanna Diagram only has a yellow and a red area: the weights they lift are either too light to create any muscle gain or too heavy to create pain and injury. However, even for them, there is a right size of weights to lift that is heavy enough to hit them above after the yellow area and before the right one, triggering muscle growth but not an injury. That’s the green area. And if they do find that size of weights and lift them, they trigger antifragility and get stronger, which means that they grow the range of weights that they can safely lift. They grow their green area. ### Becoming more antifragile The Dellanna Diagrams are useful because they help you understand how you can become more antifragile. How do you grow the green area? Easy: by shifting the threshold between the yellow and the green area to the left or by shifting the threshold between the green and the red area to the right. Let’s see what that means. Shifting the threshold between the yellow and the green area to the left means that problems that previously hit you in the yellow area and were ignored now hit you in the green area and trigger adaptation and strengthening. This means you start adapting to smaller problems. Fragile organizations do not start adapting to problems until the problem is large enough to become a priority – at which point, it is usually too late. Conversely, antifragile organizations understand that problems grow to the size they need for you to acknowledge them, so they proactively surface small problems and adapt to them before it is too late. ![The Dellanna Diagram – The left threshold](/figures/dellanna-diagram-left-threshold.png) Shifting the threshold between the green and red areas to the right means that problems previously hitting you in the red area and causing failure now hit you in the green area instead, triggering adaptation and strengthening. This means making yourself more resilient to large problems. Fragile organizations overoptimize their investments, ending up with too little balance and resources to face unexpected problems. Conversely, antifragile organizations understand that you can only adapt to problems you survive and that long-term efficiency requires capping short-term efficiency. Therefore, they always keep more resources than they need and never get into situations that might become dangerous in case of unexpected changes. ![The Dellanna Diagram – The right threshold](/figures/dellanna-diagram-right-threshold.png) In the video below, I explain in greater detail what this means concretely, both for individuals and for organizations. <![CDATA[Agency is trainable]]> http://localhost:8080/posts/agency http://localhost:8080/posts/agency Wed, 10 Jan 2024 00:00:00 GMT **That second dialogue is agency.** **Having that second dialogue is an action, albeit a mental one, and like any action, it is gated by motivation.** Put another way, you only have this second dialogue if you believe it's worth having—which depends on your previous experiences with it. If someone has low agency, it's usually because, in the past, having that second dialogue led to bad experiences. You teach agency by letting your people experience that having that second dialogue leads to good outcomes. ### How to teach agency First, let me tell you how you would do it in an idyllic world. 1. **Assign one of your subordinates the smallest task requiring agency you can think of.** It has to be job-related and meaningful, but it doesn’t have to be big. In fact, the smaller, the better. For example, “Find two ways we can lower our costs and verify their feasibility.” 2. **Make it explicit that you expect them to use agency.** For example, “I expect you to tackle any unexpected problems.” 3. **Provide a concrete example illustrating the previous point.** “For example, if you do not know how to do any part of the task, I expect you to google that, and only if Google doesn’t have the answer, ask a colleague.” 4. **Get them to work on the task.** When they complete the task, acknowledge their successful outcome, specifically highlighting your appreciation for their display of agency. This will teach them not only that they have the skills to be high-agency but also that it’s easier and worthwhile. Bam, that’s it. Doing this a few times with each of your subordinates should be enough to transform 30%-80% of them into high-agency people. The problem is that, in the real world, if you ask low-agency people to do something high-agency, they might still try to do it the low-agency way. Hence, there are three things you should be paying attention to: 1. **The task must be as simple and easy as possible.** Of course, it should still be relevant to their job and require some agency, so don’t come up with a trivial task. But it should be something that ideally can be completed in no more than a couple of hours. The larger the task, the higher the chances they fall back into low-agency mode. 2. **You should be extremely explicit and specific in your request for them to be in high-agency mode while completing the task.** Give them a few concrete examples of what completing the task in low-agency mode would look like, and tell them it won’t be enough. For instance, if the task you assigned was to invite a client to a customer event you’re organizing, you can say something along the lines of, “Just inviting the client to our event is not enough; you must make sure they read and accept the invitation, and if they really cannot come, find ways to set up a later meeting with them. No excuses.” 3. **Follow up with them frequently while at the same time avoiding micromanaging them.** Ask them how it’s going. But if they face any obstacle, do not solve it for them; just encourage them and/or repeat your expectations that they will overcome it. If you follow all three points above, the chances are that you will succeed. It won’t work all the time, not with all your employees, but it will work most of the time with most of them. Keep in mind at all times the following: **Your job is to make them undergo experiences that teach them the opposite.** ### More on this I recently recorded a video on this topic. <![CDATA[10 ways to kill motivation as a manager]]> http://localhost:8080/posts/10-ways-to-kill-motivation-as-a-manager http://localhost:8080/posts/10-ways-to-kill-motivation-as-a-manager Wed, 11 May 2022 00:00:00 GMT <![CDATA[3 Catalysts For Change]]> http://localhost:8080/posts/3-catalyzers-for-change http://localhost:8080/posts/3-catalyzers-for-change Mon, 14 Mar 2022 00:00:00 GMT People require early feedback that their efforts are not going to waste. As long as they get it, they stay motivated. And as soon as they do not get it anymore, they lose motivation and disengage. For example, if you ask your people to follow a new procedure at work, and they do it but feel no difference, no better outcome, they will learn the lesson that their efforts are going to waste. The rational decision is to disengage and abandon doing the new thing. Conversely, if the procedure has some kind of early feedback (like the "mint" in the toothpaste) showing that efforts are not going to waste, then people will be more willing and motivated to use the procedure and will be more engaged in general. This is the first of three catalysts for change: give early feedback that efforts are not going to waste. ### One way to implement the first catalyzer There are three common ways to give early feedback that efforts are not going to waste. The first one is to catch your people doing something good and thank them for it. For example, if you ask your people to follow a new procedure, then spend some time at their workplace, and when you see them do the new procedure, thank them for it. It will give them a clue that their efforts are noticed and matter. ### A common mistake When I advise my clients to do the above, a common mistake is waiting too long. If you ask your people to follow a new procedure on Monday, you should have already acknowledged their good work on Monday. If you are waiting for Friday, by Tuesday, they will have learned that their efforts are not valued. Feedback must be early; ideally, immediate. ### A second way to implement the first catalyzer The second common way to give clues that efforts aren’t going to waste is to get positive customer feedback in front of those whose work caused it. If your product changed the life of a customer, everyone needs to know about it, from the janitor to the accountant – not just the product design team. Go get customer feedback, interview your best customers, and show the relevant bits to your employees who need to see it. And do it fast – before the doubt that efforts are going to waste creeps in. ### A second common mistake When I advise my clients to do the above, a common mistake is only getting feedback from customers. However, in a company, every employee is each other's customer. Everyone benefits from the janitor’s work, for example. And yet, how many provide the janitor with clues that their work is improving their lives? Get feedback from all departments, and show it to the workers who worked hard for them. ### A third common mistake Another common mistake is to approach the above too mechanistically. For example, if you ask all your employees to make a video thanking someone else in the company, people might doubt the sincerity of these videos. Instead, hire a freelancer for a couple of days, provide them with a list of roles in your company, and ask them to find someone who benefits from each role and record a 30-second video interviewing them. Then, distribute the videos as appropriate. Similarly, such videos should not be too effusive; simply explaining why people’s work matters is sufficient. People do not need to feel loved at their workplace, but they do need to feel like their efforts are not going to waste. Aiming for the former will lead to a lot of insincerity and other counterproductive outcomes. Aiming for the latter is easier and more effective. Of course, the above won’t work for all companies, but for many, it could be a turning point in the engagement of their workers. ### A fourth common mistake Some managers understand the need to give people early feedback that their efforts are not going to waste. However, they immediately start thinking about ways to provide it in a scalable way – for example, a dashboard or a monthly report. Don’t. People need personal and early feedback, whereas scalable feedback tends to be impersonal (the dashboard) or late (the monthly report). Or, more accurately: you can also use dashboards and reports, but they should always complement personal and early feedback, not substitute it. ### Summarizing the first catalyst for change Just as toothpaste uses a mint flavor to provide immediate feedback that our efforts to brush our teeth are not going to waste, workers need similar early feedback that their efforts are not going to waste; otherwise, they will disengage. Managers should constantly ask themselves, "Am I providing that immediate feedback?" – or, more precisely, "Are my people receiving early feedback that their efforts are not going to waste?" It is a manager's job to consistently provide early feedback and, where appropriate, to collect personal and meaningful feedback from customers or other departments and share it with their people promptly. Common mistakes include waiting too long, only collecting external feedback, and making the feedback collection process too mechanical. Instead, strive for feedback that is early, personal, and meaningful. ## The second catalyst Have you ever wondered why our arteries divide into capillaries? That’s because our blood cells can only exchange oxygen with tissue cells that are close by. The same happens in organizations: people do not change when asked by someone from afar. CEOs are the pumping heart of companies, but they require a capillary structure that can drive change from close by: middle management. Yes, I know middle management gets a bad rap as an unproductive layer of the company. But that is true only for bad middle managers. Good ones are invaluable and indispensable for driving change. Put yourself in the shoes of a line worker. You receive a corporate-wide email from the CEO saying that there is a new way of doing things. You do not know the CEO, so you might not trust him: does he know what he’s doing? Does he have your best interests at heart? Or perhaps you trust him but do not trust that he knows how to do your job. Or you do not trust the full repercussions of the change he proposed. Long story short, workers are rationally suspicious of any change coming from someone they do not have a close working relationship with. Instead, a direct supervisor is uniquely positioned to explain the change to his team, listen to their questions, and answer them personally. And he can do so in a way that is impossible to reproduce at scale. Moreover, as we saw with the first catalyst, change requires early feedback that efforts are not going to waste, and that can only be given by someone close enough and trusted – reports take too long, and dashboards are too impersonal. Change requires trust, and trust does not scale, hence the need to drive change through a capillary structure made of middle and lower management. This does not mean that communication from the top is useless; on the contrary, it is fundamental. It just means that it is not sufficient. ### How to implement the second catalyst There are four steps to implement the principle of driving change through a capillary structure. First, involve middle and lower-level management. Cascade down the goal of driving change through supervisors, explain the rationale, and clarify that supervisors need their own supervisors to drive change. Hence the need for a capillary structure (CEO → middle management → supervisors → workers) rather than a flat one (CEO → supervisors → workers). Second, train middle and lower-level management. You cannot expect that your managers know how to communicate change. Train them on what to communicate and how to communicate it. Train them to listen for questions and to answer them without dismissing them. Train them to provide early feedback that efforts are not going to waste. Third, explicitly demand middle and lower-level management to be agents of change. Be clear and use visual examples. Describe three managers: one doing their job of driving change, one doing too little, and one doing too much – this will clarify expectations better than anything else. Set high standards: communication does not happen when you can be understood but when you cannot be misunderstood, and change happens not when it is communicated but when it is followed up with consistent action. Fourth, sustain the change initiative. Constantly demonstrate the behaviors you are demanding in others, and keep them accountable to the high standards you asked of them. The moment you stop is the moment doubt will arise that the change you requested yesterday is not relevant anymore (and therefore, any further effort would go to waste). ### To summarize Change requires trust, and trust doesn’t scale – hence, the need to drive change through a capillary structure made of middle and lower management. To achieve this, involve middle and lower management, train them, and explicitly expect them to be agents of change; then, constantly sustain the change initiative by taking action that demonstrates it is still a priority. Never allow fertile ground for doubts that efforts are going to waste. ## The third catalyst Years ago, I consulted an operations manager about a warehouse problem. Employees were constantly leaving mechanical parts and components scattered on the floor—a safety hazard and an efficiency drain. Despite repeated instructions to store components properly on shelves, his workers never complied. The root problem was that the manager was too busy and the warehouse too large for him to consistently enforce the standards he set for his employees. Therefore, we decided to narrow the scope of change to a single point in the warehouse, small enough for him to consistently provide feedback on. He chose the area next to a safety exit. He held a stand-up meeting with the warehouse employees and told them that the area next to the exit had to be kept clear of components at all times. He asked them to clear it immediately, and he did not leave the warehouse until it was done. Then, the hard part began. For the next month, he had to visit the warehouse multiple times a day. First thing upon arrival, he would check the floor next to that safety exit to see if it was clear. If not, he was to immediately stop whatever he was doing, walk to the nearest employee, remind them that the area had to be clear, and stay there until it was. Here is what happened. After a week, the warehouse employees learned that the safety exit area was to be kept clear. After two weeks, the employees began noticing that the area next to the machine was easier to walk in, and parts stored there were easier to find. After three weeks, the magic happened. The employees began to clear the other areas of the floor on their own initiative. You see, habits require consistency, but once learned, they can easily be expanded to other areas. The bottleneck was achieving a critical mass of consistent repetitions of the desired habit within a few weeks. However, that requires time and effort from both workers and their managers. They are both busy and have limited bandwidth. The solution was to shrink the area of change from the whole warehouse to the few square meters next to the safety exit—an area small enough for busy people to focus on it frequently enough for habits to become ingrained. This is the third catalyst for change: to avoid compromising consistency, compromise on the scope of change. Do not attempt to change too many habits or too many people at once. Instead, focus on a single habit at a time, with a single team, in a single area of their work—and then be obsessive about it for the next three to four weeks. By focusing change on a small area, obsessing over it, and allowing others to see the benefits of the change, you will achieve what you couldn’t before. This is the magic of obsessive consistency. Once the workers understand that they can change faster than their manager, they will. Once they understand that good things happen when they change, they will. Good managers embrace the effectiveness of obsessive consistency, and so should you. ## To summarize Achieving consistency requires reducing the scope of change so you can focus on it intensely for a few weeks. ]]> <![CDATA[Mimetic Societies]]> http://localhost:8080/posts/mimetic-societies http://localhost:8080/posts/mimetic-societies Sat, 28 Mar 2020 00:00:00 GMT Mimetic societies describe the state of things through the use of lagging indicators, such as GDP and stock prices. Lagging indicators are called as such because those who follow them are destined to lag behind change. 16. Mimetic societies hate leading indicators, so-called because to influence them, it is necessary to lead, rather than perform rituals. 17. Mimetic societies end in mass suicide. ]]> <![CDATA[Incentives are overrated]]> http://localhost:8080/posts/incentives-habits http://localhost:8080/posts/incentives-habits Sat, 27 Sep 2025 00:00:00 GMT when you want someone to do something, you should first ask yourself if they have already demonstrated having had that habit in the past; if so, motivate them with incentives; if not, forget about incentives and work on habit-building first. **(Of course, this is just a heuristic; there will be exceptions, but in general, you will fare better assuming it's always correct and following it directly rather than adopting more complex or tailored evaluations with a higher chance of leading to wasted effort.)** ### Developing habits I have already written about developing habits in my 2020 book, ["The Control Heuristic,"](/books/the-control-heuristic) so I won't repeat myself here. However, I want to highlight that there are two types of habits, of action and of thought, and you should consider both. Habits of action are the ones we all know: brushing teeth after dinner, etc. Habits of thought are just like other habits; they are also routines or reactions to cues, except that they are mental actions instead of physical ones: for example, the mental habit of always asking yourself what's the most efficient and effective way of completing a task before you start working on it. I mention this because, especially in business, too little time and effort is devoted to developing mental habits, despite them having a disproportionate effect on effectiveness for certain roles and tasks. Thinking about problems of the type "people are doing X too much" or "people are not doing Y enough" as an incentive problem but as a habit problem will bring you several steps closer to the solution. ]]> <![CDATA[The problem with team-building activities]]> http://localhost:8080/posts/team-building-activities http://localhost:8080/posts/team-building-activities Thu, 04 Sep 2025 00:00:00 GMT The problem with most team-building activities is that they focus on nice-to-haves while ignoring bigger problems. Employees see this misprioritization as proof that leadership is detached from reality and that nothing will improve, ever. If disengagement stems from low pay, boosting morale without showing a path to higher pay only invites eye rolls: it signals pay won’t change. And if disengagement stems from cumbersome procedures, trying to “improve engagement” without streamlining work does the same: it signals work won’t change. The same applies to any source of frustration: activities that target anything other than the biggest pain point only reinforce the belief that what matters won’t change. The bottom line is as follows. **If you run team-building events, build them around fixing employees’ main frustrations. Or skip them and improve teamwork by fixing everyday interactions** (e.g., project meetings or how managers give feedback). **But don’t run cringe team-building events that sweep structural problems under the rug.** ]]> <![CDATA[The Key To Success]]> http://localhost:8080/posts/the-key-to-success http://localhost:8080/posts/the-key-to-success Tue, 20 May 2025 00:00:00 GMT Unless you spend at least part of your time working on important-yet-not-urgent activities, you will find yourself unable to grow beyond a certain point. But why is it so hard to work on the important-not-yet-urgent? The naive answer would be that, similarly to the marshmallow test, it’s about delayed gratification. However, unlike the marshmallow test, you are not alone in the experiment room. Instead, all around you, others are eating the marshmallow, and the media are celebrating those who eat it the fastest. It turns out that the hard part is not delayed gratification but the feeling of falling behind. In particular, the hard part is the feeling that, if today you didn't get out of the day as much as someone else did, then you're falling behind, and you should catch up as soon as possible. This is a dangerous belief because it will make you feel bad every day you don't produce as much as others, and every day you don't party or vacation as hard as others. As a result, you will feel like you have no time to work on the important, be it important for your professional life or your personal one. If this is your bottleneck, here are two things you can do about it: Carefully select who you spend time with or follow online. Remove those who constantly make you feel like you're falling behind, leaving you frantically trying to catch up while neglecting the important-yet-not-urgent. Whenever you work on something important-yet-not-urgent, such as building skills, relationships, or trust, immediately give yourself a big pat on the back. It is crucial that you feel like you have made progress; otherwise, you will revert to spending all your time on the urgent. By all means, none of these are necessary. But if you examine how you spent the past week and notice that you neglected the important-yet-not-urgent, you might want to try putting them into practice. ]]> <![CDATA[How to kill a country's education system]]> http://localhost:8080/posts/how-to-kill-a-countrys-education http://localhost:8080/posts/how-to-kill-a-countrys-education Thu, 10 Apr 2025 00:00:00 GMT ]]> <![CDATA[The trajectory]]> http://localhost:8080/posts/the-trajectory http://localhost:8080/posts/the-trajectory Thu, 10 Apr 2025 00:00:00 GMT The core objective of any Performance Improvement Plan should be to reach a fast and fair conclusion. This means giving the employee a clear, actionable plan and quickly determining one of two outcomes: either they improve to meet expectations, or they would be better served in a different role. The keywords are "fast" and "everyone agrees." Prolonged uncertainty benefits no one—not the company, the team, the manager, or the employees themselves, who might find greater success elsewhere. The Trajectory Framework achieves this through a simple approach that creates clarity and alignment. ### Step 1: Visualize the Improvement Path During a one-on-one meeting, draw a simple two-axis chart on paper. The horizontal axis should represent time, and the vertical axis should represent performance. Mark two points: 1. The origin, representing the employee's current performance level. 2. A point in the upper right, indicating where they need to be in three months. Connect these points with a line—this line represents the improvement trajectory they need to follow. ### Step 2: Set Two-Week Milestones

On your chart, find the point that represents "two weeks from today." Move upward from there until you intersect the trajectory line and mark that point. This represents where the employee should be in two weeks to stay "on track."

Specifically and objectively define what behaviors, skills, and outcomes they need to demonstrate by this checkpoint.

### Step 3: Regular Check-ins and Adjustments After two weeks, meet again to assess progress against the defined milestones: - If on track as per the trajectory: acknowledge progress, celebrate small wins, and set the next two-week milestone. - If not on track: collaboratively identify barriers and develop specific actions to get back on track within the next two weeks. Continue this cycle of bi-weekly check-ins. By the second or third meeting, it becomes clear to everyone whether the three-month objective is achievable – or whether the worker simply isn't a good fit for the role. ### Notes and recommendations Some countries have specific regulations regarding Performance Improvement Plans, so consult with your HR department before implementation. You may need to adapt the timeframes or documentation to meet legal requirements. That said, try to maintain the core principles: a continuous trajectory instead of a step function, short-term milestones instead of long-term ones, and regular check-ins. ### Frequently Asked Questions **Q: How do I define milestones?** A: Ask yourself, what do I need to be able to observe in a couple of weeks to know that the worker is on track to become the kind of employee I want on my team? From a people-improvement perspective, milestones do not have to be quantitative. Qualitative ones are also acceptable as long as they are specific and concrete (for example, “improve at sales” is too generic, but “learn how to ask good questions” is specific enough). However, check with your HR partner if there are specific requirements. **Q: How do I avoid employee resistance or defensiveness?** A: Position the framework as a collaborative tool for success rather than a punitive measure. Avoid stating the need for improvement as coming from you (which can sound arbitrary), but instead explain that it is required by the job and necessary for the team/organization/mission. Emphasize that clarity benefits them by providing a fair opportunity. **Q: Where do I find the time for fortnightly check-ins?** A: These meetings do not need to be lengthy; 15 minutes is often sufficient. That said, remember that if you do not conduct these meetings, you save 15 minutes but will have to spend much more time compensating for your workers’ underperformance in other ways. ### Conclusion: From Steps to a Trajectory By shifting from a vague improvement process to a clear trajectory with visible milestones, you transform an uncomfortable situation into a structured path forward. This approach helps you have more productive conversations, create actionable improvement plans, and reach fair conclusions quickly. ]]> <![CDATA[The simpler way to get your team to use AI]]> http://localhost:8080/posts/ai-faster-adoption http://localhost:8080/posts/ai-faster-adoption Sun, 16 Mar 2025 00:00:00 GMT In all the examples above, AI is used not to replace a worker or automate a task but rather to help workers be more efficient, effective, or reliable—and to improve their skills. In 2025, I believe this is where the lowest-hanging fruit lies. Let’s see a few more high-value applications: - **Decision Support:** Helping analyze options against criteria, identify blind spots, and expand consideration of variables and consequences. - **Process Documentation:** Creating and updating standard operating procedures, training materials, and process documentation. - **Personalized Development:** Generating customized learning paths, practice scenarios, and feedback—essentially functioning as an on-demand professional coach. The above are just some of the many possible use cases. Soon, more will become viable as AI assistants add capabilities and reduce error rates. As Tyler Cowen puts it, "This is the worst they will ever be." ### Fighting resistance to change and getting AI adopted by your team Here are a few common mistakes I see managers making: - **Abstract communication.** Instead of presenting AI as an all-powerful tool with generic benefits, focus on specific, immediately relevant use cases customized to your team's needs. Explain how AI can be helpful to them, not to their manager or organization. - **Adversarial framing.** If you say or imply, “I, the manager, need you to abandon doing things your way and adopt this new method,” you inadvertently frame the conversation as you versus me, causing defensiveness. A better approach is to frame the conversation as “you and me against an external enemy,” which could be an excessive workload, competitors getting ahead, etc. - **Trying to change too much at once.** If you try to get your people to use AI for too many tasks, you won’t be able to cover one well enough to convince your people it can be used successfully. It is much better to focus on them adopting AI for a single use case and ensuring that it works so well that they will be eager to use it more. ### Rules and Personal Judgement Rules and procedures are useful, but especially when it comes to AI, you also need to cultivate your people’s judgment so they can spot eventual errors or navigate unforeseen scenarios. Thankfully, doing so is much easier than commonly thought: train your people not just by giving them procedures but also by running hypotheticals. ### Hypotheticals The fastest way to build experience and expertise within your team is to schedule a 30- to 60-minute session where you will do the following: 1. List 5-10 common and uncommon scenarios, problems, and dilemmas your team might encounter while performing a specific task. (To save time, prepare this list before the session.) 2. Present the first scenario and ask your team what they would do in that situation. 3. Provide feedback, such as “Good idea, because…” or “That does not consider that…” (It is critical that you thoroughly explain your reasoning for both positive and negative feedback.) 4. Move on to the next scenario, problem, or dilemma. This exercise allows you to provide your team with months of experience in just a few minutes. Repeat it as often as necessary. Hypotheticals are also an excellent way to prepare your team to identify potential AI errors or hallucinations. ### The importance of the first impression There is nothing worse than giving your team access to a powerful tool without explaining how to use it effectively; they will try it once, it won't go particularly well, and they will lose interest. Instead, it is paramount that their first experience goes smoothly and yields useful outcomes. Do your homework and make sure you pick a relevant use case for them to use the AI tool. Give them a few examples, show them a few good prompts, and coach them through their first use of AI. Ensuring they have a great first experience is key to effective and engaged adoption. ### Conclusions Of course, there is much more to say and do regarding how to identify tasks where AI assistants can help, how to get your people to use them, and how to train them to spot eventual AI errors and hallucinations. I plan to write more about this in the future. For the moment, let me repeat some core principles: - In addition to looking for complex and structural ways to integrate AI into your organization’s processes, also look for simple ways in which individuals can get AI assistance on their tasks and skills development. - When introducing AI, do not discuss AI tools generically but lead with specific use cases. - When training your team, after you give a couple of examples on how to use AI assistance, use hypotheticals to provide them with experience and increase their personal judgment. Training your team on using AI assistants is important not just for the benefits it can yield today but also because it prepares you and your team with the skills and experience required to leverage the inevitable improvements in AI tools that will become available soon. ### Want to discuss AI adoption further? If you are a leader, manager, or supervisor, and you either already adopted AI tools or are looking to, I’d love to chat with you. You can reply to this email or use [​this link​](https://calendly.com/lucadellanna/ai-adoption-chat-with-luca) to schedule a short call. ]]> <![CDATA[How to Use Soft Metrics to Develop Your Team's Skills]]> http://localhost:8080/posts/soft-metrics http://localhost:8080/posts/soft-metrics Sat, 15 Mar 2025 00:00:00 GMT if a player struggles with scoring, setting a hard metric like "points per game" only helps if their limitation is motivation. If their shooting technique needs improvement, this metric may actually encourage counterproductive behaviors, such as taking poor-quality shots to meet their target. Conversely, a soft metric like "shot selection quality," discussed frequently after games and practices, addresses the underlying skill gap and leads to sustainable improvement. ## How to Set and Evaluate Soft Metrics Effectively Soft metrics require more definition than hard ones. When you set a hard metric like "$5M in revenue," the meaning and measurement are clear. Conversely, when you set soft metrics, you cannot assume that your delegee knows what a good outcome means. For example, what does “running effective meetings” look like? The key is to give examples of what the good and the bad would look like. What does it look like when meetings are run ineffectively? And what does it look like when they are run effectively? Even better, I like to use my “bad/good/great” framework, which also includes what great performance would look like. Finally, avoid the temptation to measure soft metrics using hard outcomes (e.g., measuring meeting effectiveness by project completion rates). While convenient, this approach fails to develop the specific skills that need improvement. ## Examples of Effective Soft Metrics Here are concrete examples of soft metrics with the bad/good/great framework: **Meeting Effectiveness:** - **Bad:** Meetings run over time, lack clear agendas, participants are unprepared, decisions are unclear - **Good:** Meetings start on time with a clear agenda, relevant people attend, decisions are documented - **Great:** Meetings achieve progress, and people look forward to attending them, as opposed to seeing them as a distraction from their core job **Communication Quality:** - **Bad:** Emails are unclear, require multiple back-and-forths to understand, use jargon without explanation - **Good:** Messages are clear and concise, key points are highlighted, context is provided when needed - **Great:** Communication anticipates questions, provides all necessary context upfront, uses the appropriate medium for the message **Problem-Solving Approach:** - **Bad:** Jumps to solutions without understanding the problem, doesn't consider multiple options - **Good:** Defines the problem clearly, considers a few alternatives, explains reasoning for chosen solution - **Great:** Validates assumptions talking with the people directly involved, deeply understands root causes, evaluates trade-offs systematically, anticipates second-order effects ## Conclusion Complement your hard metrics with carefully selected soft metrics to support your team's development. Provide concrete examples of what bad, good, and great performance looks like for each soft metric. Most importantly, discuss performance frequently in low-pressure settings rather than saving these conversations for formal reviews. Implement this approach consistently, and within a few months, you will see both soft and hard metrics improve as your team develops new capabilities and greater effectiveness. ]]> <![CDATA[The Manager's Role from the employees' point of view]]> http://localhost:8080/posts/the-managers-role http://localhost:8080/posts/the-managers-role Mon, 07 Oct 2024 00:00:00 GMT Managers proactively communicate the hidden purpose of tedious tasks, understanding that purpose is what distinguishes tedious work from demotivating work. 7. Managers communicate so clearly and concretely that their meaning is understood by everyone. They do not clarify misunderstandings but prevent them. 8. Managers find small but value-adding tasks for their demotivated employees that they can succeed at. 9. Managers respect their employees' time. While requiring overtime may occasionally be necessary, managers take proactive steps to avoid unnecessary overtime. 10. Managers do not give up easily on demotivated employees, understanding that demotivation often stems from learning that efforts are in vain. Therefore, they find ways to direct their people's efforts toward tasks that will teach them the opposite lesson. 11. Managers provide motivated employees with opportunities to apply their motivation effectively. 12. Managers interpret abstract objectives, telling each worker what concrete actions to take and why. They never mention company-wide objectives and core values without immediately explaining what they mean concretely to the individuals they are addressing. In summary, managers prioritize, delegate, assist, explain, acknowledge, teach, and coach to ensure that their employees' time and efforts at work are worthwhile (in terms of salary, growth, self-respect, security, or whatever resource the employee joined the company for). ]]> <![CDATA[Five telltale signs of good managers]]> http://localhost:8080/posts/five-marks-of-good-managers http://localhost:8080/posts/five-marks-of-good-managers Mon, 02 Sep 2024 00:00:00 GMT employees do not take seriously leaders who do not dig deep ("How can he know about my job if he spends all his time in meeting rooms?"). ### 2. Good managers front-load clarity They do not wait for misunderstandings to happen to clarify what they mean. They are proactively clear. They do this not only because misunderstandings are a waste of time, energy, and often money, but also because misunderstandings, even when resolved, become precedents that lower the team’s trust in their manager. ### 3. Good managers explain all decisions One reason is to build skills. It is one of the most effective and time-efficient ways to increase your team's understanding of the business. However, a more important reason is that managers who do not explain their decisions might appear irrational to their team. This destroys trust that their manager is a leader worth following. Similarly, businesses whose managers do not explain business decisions to their team might appear irrational or incompetent, which also destroys trust that they are worth working for. ### 4. Good managers play iterated games with their team They do not delegate merely to get tasks done, but also to develop their team members' skills. They do not give feedback solely to correct mistakes, but also to foster trust and engagement. They do not view interactions with their team as isolated events, but as a series of interactions, where each is played not to maximize immediate outcomes, but to improve subsequent interactions. Good managers recognize trust and skills as compounding assets and continually use every opportunity to build them within their team. ### 5. Good managers treat good performance with negative externalities as bad performance Do employees take shortcuts, game metrics, violate core values, or destroy trust rather than build it? Good managers dig deep (point #1) into how people achieve objectives, and do not consider an objective achieved unless it is achieved in a way that strengthens the company. Moreover, hearing that you won’t get a bonus despite achieving your objective is frustrating. Hence, good managers are superclear since the beginning that only objectives achieved without negative externalities count (they front-load clarity, point #2). ### Further reading My two books, [Best Practices for Operational Excellence](/books/best-practices-for-operational-excellence) and [Managing Hybrid and Remote Teams](/books/managing-hybrid-and-remote-teams), dig deeper into the concrete details and best practices business leaders can use to raise their team’s efficacy. ]]> <![CDATA[Reversing the Arrow of Time]]> http://localhost:8080/posts/reversing-the-arrow-of-time http://localhost:8080/posts/reversing-the-arrow-of-time Thu, 23 May 2024 00:00:00 GMT Observe what would happen if the arrow of time reversed. We would progressively lose technology, social rights, access to services, and the abundance that we now take for granted. For example, go back in time 30 years, and we lose smartphones and the internet. Cars get worse and more polluting. Houses get smaller and are worse equipped. This effect is even worse for the bottom parts of society. Go back in time 15 years, and access to the internet disappears, but only for the bottom ~half of society. Go back in time 60 years, and plenty of rights disappear, but only for those at the bottom of society (think about racial segregation). Go back in time 75 years, and plumbing disappears, but only in the houses of the poorest 20%. Reversing the arrow of time seems evil, cruel, and a terrible injustice, not only making everyone’s life worse but especially worsening the lives of the bottom layers of society. Why do we say that the world got worse, then? ]]> <![CDATA[Races to the bottom]]> http://localhost:8080/posts/races-to-the-bottom http://localhost:8080/posts/races-to-the-bottom Fri, 11 Aug 2023 00:00:00 GMT Conversely, participants with a broad vision for success are more eager to have a full life than to win any race to the bottom. As a result, they set ambitious objectives for all important parts of their life. Then, they use these objectives to set their boundaries – what they aren't willing to risk and sacrifice. And finally, they use these boundaries to decide how to compete in races to the bottom and to what extent. These are the real winners – not in the sense that they get number one (it happens rarely), but in the sense that they get out of the race more than they risk and sacrifice. They choose the conditions at which to compete and therefore end up with a good bargain. ]]> <![CDATA[Mixed Signals, Mixed Results]]> http://localhost:8080/posts/mixed-signals-mixed-results http://localhost:8080/posts/mixed-signals-mixed-results Mon, 05 Jun 2023 00:00:00 GMT This principle surely applies to individuals. For example, those with mixed feelings about money have trouble taking decisive action to improve their income. But this principle also applies to business. Managers who give their teams mixed messages get mixed results. When they ask for feedback, they should demonstrate that they really want to hear what their team has to say. When they assign a priority, they should explicitly address that everything else is less important. And when they assign a task saying it's important, they should follow up as if it were. ### Mixed signals are the #1 reason change initiatives fail Whenever a manager introduces a new process, their team likely wonders: "Is this going to stick, or will it be abandoned after a few weeks?" To answer this, employees observe their manager's behavior. Is the manager following up on the new process and treating it as important? Or are they acting as if they don't care? Hence, it is critical that managers send clear signals that the new processes are important and here to stay. Otherwise, they might as well not try and not waste anyone's time. ]]> <![CDATA[Ergodicity as a non-binary property]]> http://localhost:8080/posts/ergodicity-non-binary http://localhost:8080/posts/ergodicity-non-binary Tue, 28 Feb 2023 00:00:00 GMT ### The purpose of studying ergodicity Ergodicity is one of the most important concepts in economics. If you are unfamiliar with the term, I suggest [my book](/books/ergodicity) on the topic. If, instead, you are familiar with the term, you might have wondered how it applies to everyday life. In particular, you might have wondered which activities in life are ergodic and which are non-ergodic. The answer is simple. Almost everything in real life is non-ergodic. The fully ergodic mostly lives only in theory and simulations. What's the purpose of studying ergodicity, then, if nothing is ergodic? The answer is that, while everything is non-ergodic, something is more non-ergodic than others - and questions such as "what alternative is less non-ergodic" or "how can we make this activity less non-ergodic" matter. I recently published [a paper](https://luca-dellanna.com/wp-content/uploads/2023/02/Ergodicity-as-a-non-binary-property.pdf) on ergodicity as a non-binary property, where I examine these questions. You will learn how to interpret ergodicity as a spectrum and how to apply it to real-life conditions. ]]> <![CDATA[How to coach your team to write more effective emails]]> http://localhost:8080/posts/email-coaching http://localhost:8080/posts/email-coaching Mon, 20 Feb 2023 00:00:00 GMT Coaching your people to write better emails is a high-leverage activity: with 15 minutes of your time, you can save them workdays writing unnecessary words, decrease wasteful back-and-forths, and improve your teams' persuasion if they deal with external clients. The best way to improve your team’s email-writing skills is to coach them individually. Focus on a single area of improvement, be very specific about why it's a problem and how they could improve it, give them actionable next steps, and validate their improvements. ]]> <![CDATA[The Lindy Effect]]> http://localhost:8080/posts/lindy http://localhost:8080/posts/lindy Sun, 06 Dec 2020 00:00:00 GMT ### What justifies the Lindy Effect? The older something is, - the more conditions it must have been fit for, - thus, the broader range of possible futures it is fit for, - thus, the longer it is likely to survive, (in the absence of bounds such as senescence). Taleb also presented a statistical justification for the Lindy Effect in his books. I won’t cover it here, as I try to keep the understanding of the Lindy Effect intuitive. ### Perishables and non-perishables The reasoning above doesn't apply to people, as senility poses a natural limit to the maximum age they can reach. An 80-year-old person cannot survive another 80 years. The Lindy Effect mostly applies to entities with no natural boundaries to life expectancy, such as technologies and ideas. For example, it applies to books, movies, and technologies like bicycles (but not necessarily to objects subject to decay, such as a bicycle). However, the applicability of Lindy based on the criterion of perishable/non-perishable is not as clear-cut as it seems. For example, Lindy doesn't apply to adults but does apply to babies. A baby that survives its first week has a considerably longer life expectancy than a newborn. Therefore, we can say that the Lindy Effect applies to perishables, but only when they are far from natural limits such as senility. As an entity approaches its natural limits, decay dominates Lindy (more on this later). ### The Hazard Rate For non-perishables, such as objects and ideas, the main determinant of life expectancy is the hazard rate (the chances of dying/disappearing at age X). When we observe an object’s life, we can use Lindy to estimate its life expectancy or hazard rate. For example, we can estimate a book’s life expectancy on the bestsellers’ list (its life expectancy) or its chances of dropping off next week (its hazard rate). Of course, the two are negatively correlated. That said, we can reason the following. The older something is, - the more conditions it must have been fit for, - and thus the broader range of possible futures it is fit for, - and thus the lower its hazard rate. Our estimate of an entity's hazard rate decreases as time passes without that entity disappearing. The first keyword is "an entity's." A book staying for months on the NYT bestsellers' list does not mean that all books on it are less likely to drop off next week. It just means that that specific book is less likely to disappear. The second keyword is "our estimate." The book's hazard rate does not decrease over time; its hazard rate is probably constant. Instead, it is our estimate that decreases. The longer the book survives, the more reasons we have to lower our hazard rate estimates. ### The hazard rate for perishables We previously saw that Lindy applies to perishables, but only when they are distant from natural limits, such as senility. Now that we know about the hazard rate, let’s clarify this sentence. We can decouple the effects of Lindy and of decay into multiple hazard rates that we can aggregate together to obtain an entity’s total hazard rate. For example, a person’s total hazard rate is made of: - The hazard rate from accidents (subject to Lindy; the more a person survives, the more we can suppose them to be cautious, and thus, the lower our estimate of their hazard rate from accidents). - The hazard rate from illnesses and internal conditions (e.g., stroke) is a component not influenced by genetic causes (this increases linearly or exponentially with age). - The hazard rate from illnesses and internal conditions is a component influenced by genetic causes (subject to Lindy – the more a person survives, the less likely they are to have genetic conditions). The total hazard rate of a person is the sum of the three points above. The second one becomes dominant as one person approaches the natural limits of human longevity. Hence, it’s not that Lindy does not influence the life expectancy of perishables – it does, but it loses relevance over time. ### The Lindy Effect, generalized Lindy applies not just to time, but also to other dimensions: space, cultures, uses, conditions, etc. Here are a few examples of practical applications. Continuing the NYT bestseller example, a book sold in one country might be successful because it’s a great book or because it discusses something very relevant to that country. Once it’s translated and does well in another country, the odds that it’s a great book increase. In general, the more geographically widespread something is, - the more conditions it must have been fit for, - thus the broader the range of conditions it is fit for, - thus the lower the estimate of its hazard rate upon entering a new geography. I suppose the same works across cultures, use conditions, and most dimensions. (Remember the limitation that “estimates made by the Lindy Effect are subordinate to intrinsic limits.” For example, a book read in 150 countries is not likely to be read in 150 more countries if there are only 200 countries on Earth.) For example, bicycles are Lindier than cars. Not only are they expected to be around for longer, but they can also be used in a wider range of conditions (off-road, in the absence of fuel) and can be built/repaired by more people with less specialized tooling. Therefore, we can often use the Lindy Effect to estimate not only life expectancy but also usefulness, relevance, and maintainability across a wider range of conditions or use cases or skills, etc. (again, a reminder: it is probabilistic, not deterministic) Before closing this essay, I have two more remarks. ### What the Lindy effect is not The Lindy Effect estimates an entity’s hazard rate, not whether that entity is good or bad. You can’t say, “It’s Lindy, therefore it’s good.” Mosquitoes are Lindy. Second, being Lindy doesn’t mean that something cannot disappear tomorrow. It only suggests we have reason to believe it is less likely to disappear than if it hadn’t been around for so long. The Lindy Effect doesn’t tell you how long something will survive. It helps you estimate its hazard rate or life expectancy, both of which are probabilistic. ### Lindyness, what is it? Lindiness is the property of being Lindy, in other words, of having been around for a long time and, therefore, being expected to be around for a long time from now. It only applies to the non-perishable (e.g., ideas, book contents, technologies, songs, etc.) and carries no moral valence. Its use is to estimate whether an assumption will still be relevant over long time horizons. ### What are some examples of Lindy? Some examples of things that are Lindy: - Books - Songs - Ideas - Technologies - Recipes Some examples of things that are not Lindy: - Food - People - In general, anything with a bounded life expectancy ### Further readings I first learned about the Lindy Effect in Nassim Nicholas Taleb’s Antifragile, a book I strongly recommend. In this piece, I’ve shared some thoughts on the process behind it and how we can apply it to more use cases. Much like this essay, [my book](/books/ergodicity) on Ergodicity simplifies a complex concept related to survival, making it practical. ### Conclusions - The Lindy Effect: For ideas and technology, every year of existence increases their life expectancy. - The Lindy Effect also applies to perishables, but only when they are distant from their natural expiration. - The Lindy Effect is not deterministic but probabilistic; it does not tell you how long something will survive, but helps you estimate its hazard rate or life expectancy. - The Lindy Effect does not tell us whether something is good or bad. - We can often use the Lindy Effect to estimate not only life expectancy but also usefulness, relevance, and maintainability across a wider range of conditions, use cases, skills, and so on. ]]> <![CDATA[Wittgenstein’s Ruler]]> http://localhost:8080/posts/wittgensteins-ruler http://localhost:8080/posts/wittgensteins-ruler Wed, 20 May 2020 00:00:00 GMT ### Examples of Wittgenstein’s Ruler The awarding of the Nobel Prize in Economics to someone can imply two things: either that the recipient is very intelligent, or that the judges are very unintelligent. At the time of the award, we likely did not know which of these two possibilities was correct. Certainly, the recipient of the prize must have appeared intelligent, but we could not ascertain whether they were truly intelligent or if the judges were simply unintelligent and easily deceived. With more than one free parameter (the quality of the recipient and the quality of the judges), we do not know which one the award assesses. A similar phenomenon can be observed during the COVID-19 pandemic. In April 2020, a few studies were published on antibody prevalence in certain populations. However, we do not know if these studies are measuring the prevalence of the virus or the accuracy of the tests. ### Wittgenstein’s Ruler: a definition Wittgenstein’s ruler can be formalized as follows: ### Extending Wittgenstein’s Ruler Interestingly, Wittgenstein’s Ruler is not just about the precision of the ruler but also about its choice. For example, centralization tends to result in the choice of metrics that, regardless of their precision, only measure some of the results that matter to the general population, leading to effects such as “centralization is only efficient to the central observer.” This is because the central observer is the one who chooses the ruler, i.e., the metric used for measurement. I used to understand the term "ruler's reliability" as simply a matter of precision/variance; instead, it's also a matter of the choice of the ruler and the metrics used to conduct the measurement. Do they reliably help estimate properties of the object of the measurement, or do they estimate something else? Hence, we can use Wittgenstein’s ruler even before the measurement is conducted, using the choice of the ruler to deduce the properties of the measurer. ]]> <![CDATA[Critical Mass]]> http://localhost:8080/posts/critical-mass http://localhost:8080/posts/critical-mass Sun, 10 Nov 2019 00:00:00 GMT **The manager achieved what he couldn't before by focusing change over a small area, obsessing over it, and letting others see the benefit of the change.** This is the magic of obsessive consistency. Once the workers understand that they can change faster than their manager, they will. Once they understand that if they change, good things happen, they will. Good managers know the effectiveness of obsessive consistency. ### Good managers aim to reach critical mass They know that habits need a critical mass to form. Some experts say an action becomes a habit once it is repeated for 21 straight days. Others say it becomes a habit once everyone in the room performs it simultaneously. Both groups of experts recognize the importance of reaching a critical mass of action, both in terms of time and social impact. Consistency is essential for two reasons. First, it helps to achieve the critical mass that leads to action being converted to habits in our brains. Second, lack of consistency kills habits. The first time that you allow an employee to perform below standard without consequence, you open the door to him doing it a second time or to others doing it for the first time. They and their peers will interpret it as a signal that no one cares if someone performs below standard. It doesn't matter whether it is true. What matters is the perception that is created. Because good managers know the importance of reaching critical mass, they restrict the scope of the change to be achieved at a time. Instead of asking their subordinates to adopt five new habits this month, they ask them to adopt only one. Otherwise, they would have to divide their limited time between noticing and praising or reprimanding five habits or lack thereof. By focusing on noticing a single habit, good managers decrease the likelihood of missing an instance where a subordinate failed to express the required habit. Because they know the importance of reaching critical mass, good managers restrict the number of people who must adopt a new habit at a given time. Instead of requiring their whole building to adopt a new habit, they focus on one or two teams at a time. This way, they won’t have to spread their attention too thin. They ensure that no bad habit will go unnoticed and no good one unreinforced. They know that employees watch how their colleagues are treated. If they see a colleague failing to adopt a new habit and nothing is happening, they will learn that they can do the same. By obsessing over a single new habit of a small group of people at a time, good managers ensure that no such occurrence occurs. ]]> <![CDATA[Pain is a signal of vulnerability]]> http://localhost:8080/posts/pain-signal-vulnerability http://localhost:8080/posts/pain-signal-vulnerability Wed, 02 Aug 2017 00:00:00 GMT _Disclaimer: I am not a medical professional, and this is not medical advice. This post only states my beliefs as a result of my research on the topic. Full disclaimer [here](/disclaimer)._ ### Introduction Millions of people are affected by chronic pain. For some, an injury or a disease is the cause. For others, the pain originates from their mind. This essay explains the rationale behind this behavior of our mind and explores some ways to cure this type of chronic pain. ### The purpose of pain Pain serves a purpose. When we twist our ankle, it becomes painful. This is good, as it makes us aware of the injury and ensures that we do not step on it before it has healed. ### But pain is not a signal of damage Many believe that pain is a signal of damage occurring somewhere in our body. This is incorrect: pain is a signal that we are vulnerable, that damage can occur to us in the future. Most times, we are vulnerable because of damage (as in the example of the twisted ankle), hence the confusion between pain as a signal of damage and pain as a signal of vulnerability. I will try to explain why pain is indeed linked to vulnerability, but not (necessarily) to damage. Not all the time that there is damage in our body do we feel pain, and not all the time do we feel pain when there is damage in our body. However, every time that we feel pain, there is a potential for future damage (a vulnerability). Here are some examples of cases in which our body has been damaged and yet, pain is not felt: - There are numerous reports[^1] of soldiers in World War I and II who lost a limb in an explosion and yet did not feel any pain. Why? Because the injury meant they would not have to fight anymore and would soon be sent home. They felt less vulnerable to death in a hospital bed with a severed limb than when they had to fight in the trenches. The future felt safe, or at least safer, so no need for excruciating pain anymore. As I will show later, the pain neurons in the affected limb still fire to signal pain, but the signal is probably suppressed by their brain. Now that the soldier is on a boat on the way home, it has no utility towards changing their behavior: they are now attended by doctors and on their way home. - When participating in sports, it's possible to sustain a minor injury. Often, the pain isn't felt until after the physical activity has ceased. This is likely because sports often resemble activities like fighting or fleeing. In such situations, there's a much greater risk of bodily harm if one rests in the presence of an enemy or predator that could kill them, compared to stepping on a twisted ankle or continuing to use a sprained muscle. Consequently, using the injured body part is perceived as less dangerous than not using it, and no pain is felt (at least after the initial few seconds). The pain signal is suppressed. Examples of cases in which we feel pain but there is no actual damage to the body: - If your finger touches a hot pot, you will feel pain, even if the touch only lasted for an instant and your skin did not receive any actual damage. In this case, pain was not a signal that your skin got damaged (it didn’t), but a signal that your skin might get damaged if the behavior persists (i.e., your skin is vulnerable). Similarly, when exposed to extreme cold for a short amount of time, your skin will feel painful. It is not damaged yet, but it will be if the exposure continues. Pain is not a signal of current damage, but of future one. - Science writer Erik Vance tells us an interesting story. Sitting in a lab, he was administered electrical discharges. When the screen in front of him would turn red, he would receive a stronger charge; when it would turn green, he would receive a lighter one. After following this pattern for a few minutes, the researchers changed the rule: all discharges would be strong ones. Erik, who was not aware of the change, still perceived little pain when the screen was green, even though the actual discharges were strong. His brain overrode reality with expectations (of imminent body damage). Expecto, ergo est.[^2] - Psychosomatic pain. A condition for the emergence of psychosomatic pain is a perceived condition of generalized vulnerability (and, by association, future physical damage). I will clarify this later in the chapter. I can draw an example of psychosomatic pain from my personal experience. When I was 23, I noticed a fast-growing black mole on my right foot. I did not feel any pain, but I decided to visit a dermatologist anyway. After the examination, he told me that the mole could degenerate if left unchecked. We scheduled the removal for the following week. During the seven days between the dermatologist’s visit and the surgical operation, my foot ached. Of course, there was no good reason for my foot aching: I had no injury, and there is no way that a tiny mole, even if malignant, would be painful. However, there I was, feeling pain. Why? During the first visit, the dermatologist said that the mole could degenerate if not removed. My brain captured that information and inferred that my foot was vulnerable. To ensure that I would stay focused on the task of removing the mole, it made me feel pain. (How physiologically my brain managed to make me feel pain will be explained later.) ### Damage does not have to be physical Physical damage can predict further physical damage. For example, if our ankle is twisted, we are in a vulnerable situation. Stepping on our ankle might make the injury worse, and a twisted ankle is also a liability if we have to escape from a predator or fight an enemy. **Physical damage is a sign of a vulnerability.** However, current physical damage is not the only predictor of future physical damage. **Psychological damage, social damage, and lack of resources also predict future physical damage.** When we are healthy but in a condition that might lead to future ill health, we are vulnerable. Some examples: - If I lack the resources I need for living (food, money, shelter, etc.), I am at a higher risk of physical deterioration. - If I did something wrong which harmed other members of my community, I face the risk of being isolated and ultimately ostracized by my community (social damage). Alone, it is much harder to find the resources needed for living and the help or support in case of need: over time, physical deterioration might follow. - If I am depressed (psychological damage), I am a less attractive mate and a worse friend. This might lead me to be left alone by my previous friends. As explained in the previous bullet point, if alone, I am more likely to face a shortage of resources and external support, which directly leads to a higher risk of physical damage or deterioration. Therefore, **it makes sense for my body and brain to consider a lack of resources, social damage, and psychological damage as a vulnerability and thus akin to a risk of physical damage. This risk of physical damage then manifests as pain.** However, such vulnerability and pain are generalized: they are linked to our personal situation, but not to any specific part of the body. **If our body does not have reasons to target a specific part of our body with pain, it manifests the generalized vulnerability as stress.** In the next section I will explain this process. ### Stress as generalized vulnerability First, let me explain the purpose of stress. The feeling of vulnerability is used to trigger reactions and to find solutions to the root cause of the vulnerability (how this takes place in practice will be the topic of the next section). However, such reactions, like all actions and reactions initiated or mediated by our brain, must undergo motivational gating by the basal ganglia. If such reactions are repressed (they do not manage to overcome the motivational gating), no solution or damage mitigation plan is found, and the vulnerability remains unaddressed. What started as a clear signal of vulnerability is now a generalized signal of vulnerability. Our brain still knows that something is wrong, even if repression through motivational gating does not allow it to know exactly what is wrong. Nevertheless, something has to be done. The signal that something has to be done is stress. Our brain is mostly an inference machine. The role of each neuron or group of neurons is to recognize a pattern in the inputs it receives. Such inputs are of three types: sensorial data (bottom-up), context (lateral), and expectations (top-down). Let’s see a (much simplified) example of how this works. Let’s say that a neuron’s job is to fire when it recognizes a dog. This means that it will fire if it recognizes sensorial input corresponding to four paws, a body of a certain size, fur, a head, and a tail. If the only sensorial input is a tail, it will generally not fire. It needs a minimum number of sensorial stimuli matching the pattern of what a dog looks like in order to fire. This minimum number is called the admissibility threshold. For example, it might fire when 4 of the 5 visual characteristics of a dog listed above are recognized. However, top-down expectations might reduce the amount of sensorial data needed for the neuron to fire. If I am at home and I own a dog, I expect to see it in the living room. If, through the kitchen door, I only see a tail, my neuron will fire: it is highly probable that that tail means that my dog is there. In other words, top-down expectations reduce the admissibility threshold and thus the amount of proof (sensorial stimuli) needed to conclude that what is expected is indeed there. Now, enter stress. **Stress is a state of general vulnerability:** our brain knows that we are vulnerable, but this vulnerability is not associated with a specific body part and thus does not generate pain. However, because of the vulnerability, our brain has a higher expectation of feeling pain. This top-down expectation lowers the sensory threshold for experiencing pain. My hypothesis[^3] is that, because of this top-down expectation, our brain will be more likely to find admissible sensory signals that are precursors of pain. Let’s see an example: I do some work in the backyard. Usually, I would need to lift a very heavy load to damage my back. Let’s say that lifting 50 kg would cause my back to suffer damage (such as a herniated disc). At this point, I will suffer pain: a signal that I need to rest; otherwise, I will suffer worse injuries. Even after healing, my brain is likely to remember that lifting 50 kg might cause back damage. The next time I lift 50 kg, even if my back does not actually get injured, I am likely to feel pain: a signal that I’m vulnerable to damage. Now, let’s imagine that I am in a stressful period of my life. This time, the pain threshold will be lower (due to the stress). One of two phenomena is likely to occur: 1. Since the pain threshold is lower, the sensory signals sent by my back to my brain when I lift 40 kg are enough to cause pain. I might think that I am injured and have a herniated disc, even if my back does not actually have any. 2. A generalized vulnerability manifests as stress as long as it does not have any admissible location to manifest as bodily pain. Lifting the weight gives my brain an admissible location to feel the pain: my back. So, I feel pain there. (In some cases of chronic pain, doctors recommend serotonin reuptake inhibitors. This is probably because a lack of serotonin signals a lack of resources, which is a condition of vulnerability and therefore causes the pain admissibility threshold to be lowered.) You might feel skeptical: can our brain truly imagine pain? Why would it do that to itself? Let me show you some medical results that suggest it truly is like that. ### John Sarno’s patients In his books _“The Mindbody Prescription”_ and _“The Divided Mind”_, Dr. John Sarno describes numerous cases of patients with psychogenic pain: pain engineered by our brain. Dr. Sarno would conduct an objective analysis to eliminate other diagnoses. Then, he would interview them and notice that they were very stressed. He would then tell them that their pain was psychogenic: it originated in their brain. They did not feel pain because something was wrong in the body, but because their brain was making them feel pain. Their brain, he would explain, was doing so in order to distract them from thinking about inadmissible thoughts about themselves, which were unconsciously generating guilt, shame, and other emotions that would all create stress (Author’s note: I do not agree on this very point – I will explain my theory later). In some patients, the pain vanished over the next 24 hours; others needed to attend a few group sessions in which such mechanics would be explained in more depth. The results were surprising: after having worked with Sarno, about 85% of his patients reported improvements in their condition, with 44% of them reporting little or no pain.[^4] How could the brain of his patients generate pain? Sarno hypothesized that the brain achieved this result by contracting some blood vessels and generating mild ischemia (deprivation of oxygen) in the target tissues. If you doubt the ability of the brain to alter the size of blood vessels, just think about how we blush after doing something embarrassing: our brain increases the diameter of the blood vessels in our cheeks. ### Arthroscopic surgeries of the knee Sham surgeries are fake interventions where the patient is transported into the operating room and anesthetized. However, the doctors do not perform any surgery – they merely make the patient believe they did. The results have been surprisingly positive: for example, **sham surgeries for arthroscopic surgery of an osteoarthritic knee proved to be as effective as real ones.**[^5] This means that at least some cases of knee pain are not caused by actual body damage, but by the perception of a state of vulnerability. ### Herniated discs In his book _“The Mindbody Prescription”_, Dr. John Sarno reports a study published in the journal Spine. Doctors made lumbar CT scans of a group of people without lower back pain: they found disc abnormalities, stenosis, and other aging changes in 50% of patients over 40 years old. Such abnormalities were not causing any pain. Contrast this with the common procedure when a patient tells a doctor he has been suffering from back pain. Often, a scan of his back is ordered; if an abnormality is found, such as a herniated disc, responsibility for the pain is attributed to it. Sometimes, surgeries are even recommended. However, if herniated discs are also present in painless patients, how can we be sure that they are the cause of the pain in patients with back pain? There is a chance, writes Dr. Sarno, that at least in some of the patients with chronic back pain, the herniated disc is not the cause of the pain (or it is the cause of its onset, but not of its persistence). In such cases, psychogenic pain would be the cause. ### The sources of pain There are three sources of pain: - **Nociceptive pain:** This is the pain we generally refer to in common conversation. It is caused by external harmful factors that excite our nociceptors (the neurons responsible for detecting extreme heat, extreme cold, wounds, impacts, and so on). This type of localized pain causes a bottom-up inference of a localized vulnerability. - **Psychosomatic pain:** This pain occurs when nothing is wrong with our body, and yet our brain infers from our present situation that we are generally vulnerable (e.g., to social isolation, lack of resources, etc.). This inference generates stress and a top-down expectation of pain, which is eventually manifested in the most admissible body location. - **Psychogenic pain:** Technically, this is a concurrence of the two previous cases. However, I have listed it as a separate occurrence to highlight the specific process. As with psychosomatic pain, our brain expects a body part to manifest pain. Different from purely psychosomatic pain, this top-down expectation triggers some physiological processes (such as mild ischemia – a lack of oxygen delivered to muscles[^6]) which in turn trigger nociceptive pain. In this case, there is something unusual with our body, but only because of signals originating from our brain, causing the unusual condition in our tissues. Some may be skeptical that our brain can indeed generate changes in our body that, in turn, provoke pain. But just consider how our cheeks turn red when we are embarrassed: our brain can very easily initiate such physiological processes. Some might be skeptical that our brain indeed can generate changes in our body which in turn provoke pain. But just think about how our cheeks turn red when we are embarrassed: our brain can very easily initiate such physiological processes. ### Chronic pain Many theories on chronic pain propose that its onset is triggered by an episode of extreme stress. I believe that this is partially true. My hypothesis is that **stress is not the trigger of chronic pain, but the enabler.** There would be little reason for pain to persist if the vulnerability (the stress) is gone. Rather, it makes a lot of sense that pain persists as long as the vulnerability is there, and that it goes away once the vulnerability is gone. ### Placebos Professor Nicholas Humphrey tells a story[^7] about placebos and hamsters. “Suppose a hamster is injected with bacteria, which makes it sick – but in one case, the hamster is in an artificial day/night cycle that suggests it’s summer; in the other case, it’s in a cycle that suggests it’s winter. If the hamster is tricked into thinking it is summer, it throws everything it has got against infection and recovers completely. If it flings to study its winter, then it just mounts a holding operation, as if it is waiting until it knows it’s safe to mount a full-scale response. […] In winter, we are conscious about deploying our immune resources. That’s why a cold lasts much longer in winter than it does in summer. […] Placebos work because they suggest to people that the picture is rosier than it truly is. […] Placebos give people fake information that it’s safe to cure them.” The purpose of pain is to modify our behavior to reduce our exposure to the vulnerability causing the pain. This might include finding a solution (e.g., treating a wound), distancing ourselves from the potential source of harm (e.g., taking our finger away from a hot pot), and putting ourselves in a situation where it is safe to heal (e.g., staying in bed). Healing is often a costly process. Knowing whether now is the time to address the root source of harm is an important ability. Our body has to commit resources (energy, nutrition, attention, time, antibodies, etc.) that would be better used elsewhere. Here are some examples of situations where it is not a good idea to heal: - A wolf bites my hand. Instead of treating the wound, I should try to scare off the animal and then seek medical attention. - The harvest is late, and we are suffering from malnutrition. Instead of lying in bed to reduce energy consumption, we should work in the field to ensure we have the maximum number of vegetables once they are ready. - I sprained a leg muscle during a long hike alone in the mountains. The optimal reaction is to move cautiously to reach home, where it will be safe to rest and heal. Our brain evolved the ability to infer whether now is a good time to commit the resources required for healing, or to persist in our default behavior unmodified by pain. Many factors can influence this inference. In particular, stress (which is a signal of a generalized vulnerability) might suggest to our brain that now is not the time to heal, and that it is appropriate to feel pain is a signal that we are not safe yet, and a reaction is needed. **Placebos work by suggesting to us that we are safe. If we are safe, there is no longer a need to conserve resources for fleeing or addressing the external source of harm. Instead, we can now commit them to healing.** ### Behavioral placebos In the previous paragraphs, I said that placebos are suggestions that we are safe and can commit resources to heal. Placebos are permissions to heal. There is a second definition of placebo, which is more interesting from a behavioral point of view: **Placebos are permissions to change.** They are narratives we can tell ourselves to justify a change in our behavior. Here is a story to use as an example: _Elbert always wanted to dress more elegantly, but he never did so because it “wouldn’t fit him.” He already had some suits in his wardrobe and liked to wear them during weddings and other formal events. However, he could not bring himself to wear them on any other occasion. He feared that others would ask questions such as, “Why did you start dressing elegantly?” and “Why didn’t you do it before?” These thoughts prevented him from dressing up for a long time. One day, he received permission to change: his wife gifted him a suit. Finally, he had a coherent narrative to justify wearing one on informal occasions. It would not be his choice: it would be his wife’s._ Comedians not only know a large number of funny jokes; they are also very good at giving us permission to laugh. Similarly, a lady might have different perceptions of a serenade, based on whether it is performed by a cool, handsome guy or by a shy and uncool one. In the former case, it is perceived as a romantic gesture; in the latter, as a creepy one. Rory Sutherland said, “Trumpets and marching are bravery placebos.” Placebos allow us to be confident. (As a side note: if being confident is a good thing, why aren’t we all always confident? It is because being confident isn’t always a good thing. Often, it is a bad idea to be confident when there are no reasons to be confident. For example, it can lead to being perceived as arrogant, as a bully, and lead to shame and ostracization. This is why our brain had to evolve the ability to infer from the situation at hand when to be confident and when not to. Lack of confidence, like all bad feelings and emotions, has an overall beneficial purpose or is the necessary byproduct of something beneficial.) ### Conclusion Physical damage is only one of the causes of pain. If you suffer from chronic pain with no clear physiological cause or with a physiological cause that doesn’t seem to heal, consider the possibility that your unconscious self might feel so threatened that it believes that pain is an appropriate reaction. In that case, two approaches might be beneficial: placebos, and taking care of those sources of stress that are making your unconscious self feel vulnerable. _Note: this was an excerpt from the 2nd edition of my book, [The Control Heuristic](/books/the-control-heuristic)._ #### Footnotes: [^1]: As reported in Henry K Beecher’s studies. [^2]: Some readers might still doubt our mind's ability to consider inferred stimuli equal to sensed stimuli. The paper “Pavlovian conditioning–induced hallucinations result from overweighting of perceptual priors” (Powers, Mathys, Corlett, 2017) reports that “Pairing a stimulus in one modality (vision) with a stimulus in another (sound) can lead to task-induced hallucinations in healthy individuals. After many trials, people eventually report perceiving a nonexistent stimulus contingent on the presence of the previously paired stimulus”. [^3]: The next part of this sub-chapter on pain is based on the work of Dr. John Sarno. I personally quite agree with his theories regarding how psychogenic pain is generated and how to cure it. I think, however, that he missed part of the purpose of pain: it is not (only?) a distraction, but a signal of vulnerability (and thus, a desperate focus signal). I also think he failed to catch the similarity between stress and pain: they are two faces (generalized and localized) of the feeling of the same concept: vulnerability. [^4]: John Sarno, “The Divided Mind”. [^5]: Moseley JB, O’Malley K, Petersen NJ, Menke TJ, Brody BA, Kuykendall DH, Hollingsworth JC, Ashton CM, Wray NP (2002). “A controlled trial of arthroscopic surgery for osteoarthritis of the knee”. New England Journal of Medicine. [^6]: Dr. Sarno named TMS (Tension Myositis Syndrome) the clinical condition whose symptom is psychogenic pain induced via mild muscle ischemia. [^7]: In his edge.com piece dated 2011-05-12 [^8]: In Farnam Street’s podcast with Rory Sutherland ]]> <![CDATA[Hercule Poirot and solving business problems]]> http://localhost:8080/posts/quality-information http://localhost:8080/posts/quality-information Sat, 20 Sep 2025 00:00:00 GMT Quality information can only be collected through methods that do not scale: in-person conversations or video calls, where the interviewer can ask follow-up questions and dig deep to capture the full context. If you are a busy executive, you might lack the time to collect high-quality information and wish to delegate it. That is fine, provided you follow these guidelines: - **Allow only methods that do not scale.** No surveys or data reports, only in-person or video interviews. At least 75% should be one-on-one: group interviews have benefits, but social pressures limit honesty and prevent hard questions. - **Do not let your delegee delegate the interviews further.** The more people a piece of information passes through, the more a “broken telephone” effect corrupts it. - **No summaries, only full anecdotes.** Your delegee may filter for importance or cut irrelevant parts, but never summarize. The value is in the details. - **Tell your delegee to dig deep, relentlessly asking “why.”** Why did this happen? Why did they do this instead of that? What is their guess about someone else’s actions? - **Instruct them to collect pieces of the puzzle, not solve it.** Guessing solutions too early risks misleading or halting the investigation. First, gather information, then solve the puzzle. - **Preserve names whenever possible.** Privacy matters, so get permission to use names whenever possible. This will be invaluable for follow-ups. Moreover, anecdotes with a name and a face are easier to remember and more likely to drive change. ### Two takeaways 1. Do not jump into problem-solving mode before having collected enough pieces of the puzzle. 2. It's often a better use of your time to dig truly deep into a single anecdote than to collect more superficial data. Only the former will help you truly understand what's going on and why. ]]> <![CDATA[Positive-sum games are the optimal selfish choice]]> http://localhost:8080/posts/positive-sum http://localhost:8080/posts/positive-sum Tue, 19 Aug 2025 00:00:00 GMT The point is: positive-sum effects dwarf zero-sum ones over the long term. This is somewhat evident. What's less evident is that the same game can be played with zero-sum or positive-sum tactics. Think about the competition between companies. Do they compete mostly by focusing on zero-sum tactics or positive-sum ones, such as technical innovation? Even customer acquisition in the absence of innovation can be zero-sum or positive-sum, depending on whether the tactics used increase trust in the product category or decrease it, for example, by lying to customers or breaking their trust. Or think about how countries fight poverty: do they do it predominantly using zero-sum tactics, such as financial redistribution (which redistributes scarcity), or positive-sum ones, such as building the capacity to produce enough goods and services so that everyone can get some (creating abundance)? Or think about higher education. Are degrees designed and attended predominantly for their zero-sum benefits (a piece of paper that gets you ahead in a zero-sum job search) or their positive-sum benefits (useful skills that increase the country's prosperity)? But what determines if society will play a given game using zero-sum or positive-sum tactics? Plenty of factors and incentives matter, sure, but an underrated one is the belief of whether one's prosperity mostly depends on zero-sum or positive-sum effects. If people believe life is mostly zero-sum, they’ll default to zero-sum tactics. If they believe in positive-sum, they’ll invest, cooperate, and innovate. And, perhaps most importantly, treat societal trust, societal infrastructure, and societal productive capacity as a common good, worth building and never worth consuming. The tricky part is that, indeed, using zero-sum tactics generally gets you ahead in the short term, but if you want your quality of life to grow beyond a certain level, you cannot do without society predominantly using positive-sum tactics. This is, I believe, the key point we should hammer. It's paramount that people understand that playing positive-sum games is not only the optimal choice from a prosocial perspective but also from a selfish one. ]]> <![CDATA[The missing element in change initiatives]]> http://localhost:8080/posts/the-missing-element http://localhost:8080/posts/the-missing-element Sat, 16 Aug 2025 00:00:00 GMT if you want a change initiative to be effective, you must send strong, clear, and unambiguous signals that the initiative is here to stay. Moreover, you must ensure that every single manager and supervisor also sends strong, clear, and unambiguous signals that the change initiative is permanent. Anything less than that, and your change initiative will likely fizzle out. ]]> <![CDATA[Screwworms and Maintenance Culture]]> http://localhost:8080/posts/maintenance-culture http://localhost:8080/posts/maintenance-culture Sat, 19 Jul 2025 00:00:00 GMT The first half of the solution is the realization that the default trajectory of large-scale maintenance projects is toward complacency and that, crucially, incentives do not change that. Do not get me wrong: incentives are powerful in most situations; just not for maintenance projects. This is because of a few reasons: - **Incentives work better on ambitious people,** but ambitious people do not want to work on maintenance projects, nor should we want them to, due to high opportunity costs. Conversely, less ambitious people have lower receptivity to financial incentives. - **Incentives work better in projects whose lead time between input and output is short.** Maintenance projects tend to have a large lead time, in the sense that maintenance quality can degrade for years without cracks showing. When incentives are ineffective, we shall turn to cultural solutions, which are largely about creating and sustaining habits that maintain complacency low and quality of execution high. Here are some examples of what this could mean in the context of maintaining the screwworm barrier: - **Proactively discuss complacency instead of hiding it under a rug.** Repeat, often, that the greatest danger is complacency. Explicitly mention the logical reasonings that might lead one to become complacent and collaboratively discuss how to prevent them. Explain that no negative event happening for years is not a reason to believe that the current level of investment in maintenance is excessive and could be lowered. - **Regularly remember the failures of the past.** There’s a reason why, in Europe, many countries have a day of remembrance where institutions share stories about the Holocaust: we must remember how bad the past was and that it might revert to being bad if we let our guard down. (How would I improve the day of remembrance? By discussing not Nazism specifically but authoritarianism in general.) Similarly, everyone working on the project must be reminded how big a threat screwworms were and still are. - Regularly remind your people that dystopia is the default, and the current utopia only exists because of past projects that must be actively maintained. Just like poverty, and not prosperity, is the default for the human race, we must remember that screwworms are the default, and their absence is only due to the hard work of a small group of people, and will cease once that work ceases. - **Regularly reinforce the good work, even when it seems unnecessary.** That’s because if people do good work and they don’t get acknowledged, they will soon stop doing it, or do it as superficially as possible. - **Take near misses seriously.** When failure is catastrophic, you cannot wait for it to happen before implementing corrective actions. Instead, you should continuously be on the lookout for near misses: incidents, mistakes, and misunderstandings that did not lead to material consequences. Take near misses as seriously as consequential incidents: investigate what went wrong and implement the necessary adjustments to ensure it never happens again. - **Do all of the above more frequently than you think is necessary.** That’s because of a natural bias called risk homeostasis that misleads us into doing less than necessary to fight low-saliency threats. - **Do it as if it were a question of life or death.** When you say maintenance is important, people will not take your words at face value. Instead, they will check whether your actions match those of a person who truly believes maintenance is important. As I said, the above represents only half of the solution. Here is the other half. ### Cultural solutions require skip-level in-person interactions Managers in large hierarchies, such as those necessary to manage a large country or corporation, tend to have managerial interactions only with their direct superiors and subordinates. This creates a game of broken telephone, where each weak link in the hierarchical chain can distort information flows. To visualize this, imagine being the coach of a sports team, but instead of talking directly to your players and watching them play, you only talk to your assistants. In this situation, the information you receive about how your team plays is only as good as the observation skills of your assistants, and the messages your team receives from you are only as good as the communication skills of your assistants. Thankfully, coaches watch their team play in person and address them in person, too. The same type of skip-level interactions should also happen within a maintenance project. If the screwworm barrier project is so important, the President should visit the team once or twice during their tenure. If this is unrealistic, at the very least, the President should speak directly about it to a delegate (say, the Secretary of Agriculture) who, in turn, should visit the screwworm team in person. Reducing the number of links from many to two means preventing the dilution of saliency, urgency, information, and recognition. _(Note that, by “visiting the team,” I mean both the person or committee responsible for managing the project, and the people working on the ground.)_ The same reasoning applies to the person whose job is to lead the project. They should “visit the grounds” to observe the operations and meet the line employees at least monthly. This is not only to communicate the importance of the project and fight complacency, but also to observe how local managers and supervisors do their work (Do they know what’s going on? Do they take near misses seriously?) and to evaluate the mood (Do people look tired or complacent?). Of course, I’m well aware that the above takes precious time, which is the resource that managers and leaders lack the most. But it is necessary to sustain good outcomes over time, especially for maintenance projects. I’m not saying to dedicate more time to maintenance projects than necessary; just an amount that matches the project’s importance, without deluding ourselves that they can succeed even in the absence of building an appropriate culture. ]]> <![CDATA[Cars, efficiency, and drivers]]> http://localhost:8080/posts/cars-efficiency http://localhost:8080/posts/cars-efficiency Sun, 07 Jan 2024 00:00:00 GMT The objective should be to find a way to offer alternatives that don't require subsidies or coercion to be chosen – that's when we will truly have something more efficient. ]]> <![CDATA[Managerial Transitions]]> http://localhost:8080/posts/managerial-transitions http://localhost:8080/posts/managerial-transitions Sat, 15 Nov 2025 00:00:00 GMT If a manager cannot correctly assess individual performance, praise those who do well, and reprimand those who do not, things will break down quickly. Good performers will feel frustrated that their efforts apparently do not matter, at least not to their coach, and will wonder why they should bother. Poor performers will notice that they can get away with subpar work, and will act accordingly. In this sports scenario, players still have incentives to perform well in the eyes of the media to land sponsorships, and in the eyes of opposing coaches to attract better contracts. However, in a company, these incentives are much weaker, especially if we consider companies whose hires are not extremely ambitious. What dominates is the incentive to perform well in the eyes of one's boss, the person with the most direct control over their bonuses and promotions. This is why one of the most important skills for any manager or supervisor is the ability to evaluate the performance of their subordinates. This is especially relevant for roles whose performance cannot be assessed quantitatively and must be judged qualitatively. If the manager cannot spot quality, their team will not produce it. Moreover, this skill must apply not only to evaluating overall performance, but also to assessing performance in each separate task. A manager who can only judge the overall performance will struggle to stimulate improvements in individual skills. One needs to be able to say, "Your overall performance is good, but you can do better at X and Y." And add, "Your good overall performance is no excuse for not improving at X and Y, because they are necessary to reach the next stage." This is why one of the key areas I work on with my clients is the following question: are you, and are the managers in your organization, able to assess how managers manage people, not only by looking at bottom-line metrics, but also by examining what they actually do? Otherwise, you will end up like a basketball coach who only praises or reprimands his team based on the final score. It seems logical, but it does nothing to improve skills. It separates the link between input and output, and leads to worse performance in the future. ]]> <![CDATA[Scientific Rigor]]> http://localhost:8080/posts/rigor http://localhost:8080/posts/rigor Fri, 31 Oct 2025 00:00:00 GMT there is nothing scientific in accepting results that contradict the basic laws of physics. (For lockdowns not to work, either germ theory would have to be wrong or lockdowns would have to fail to reduce social contact. Are these reasonable claims? A similar logic applies to masks.) Performing complex statistical analyses, drawing professional-looking charts, and packaging them into a LaTeX-style document does not compensate for a lack of basic rigor. **(\*Source: death data from NHTSA; mileage data from the U.S. Bureau of Transportation Statistics, both for 2021.)** ]]> <![CDATA[Medical Culture]]> http://localhost:8080/posts/medical-culture http://localhost:8080/posts/medical-culture Sat, 11 Oct 2025 00:00:00 GMT That's because we learned only the superficial lesson from Semmelweis's story, that doctors must wash their hands, and not the deeper one: that culture and the status quo can blind even knowledgeable professionals to obvious truths. The second lesson is so important that it should be a core point of high school education. _(Related: my essay on [​teaching tacit knowledge​](/posts/tacit-knowledge))_ ]]> <![CDATA[How transferable are best practices?]]> http://localhost:8080/posts/best-practices-transferability http://localhost:8080/posts/best-practices-transferability Sat, 04 Oct 2025 00:00:00 GMT it is crucial to define a best practice not by what it looks like, but by what it does and the principles it relies on. ]]> <![CDATA[Tacit knowledge]]> http://localhost:8080/posts/tacit-knowledge http://localhost:8080/posts/tacit-knowledge Sun, 24 Aug 2025 00:00:00 GMT Do not lead with theory; doing so is paralyzing, will make you less comprehensible and relevant, and, most importantly, will mislead you into selecting the concepts that are easier to express abstractly and neglecting the rest. Conversely, lead with concrete situations. That will help you focus on judgment calls and other concepts that are hard to explain abstractly but easy to discuss when related to a real-world situation. ]]> <![CDATA[Talking with your Manager about Better Management]]> http://localhost:8080/posts/talking-with-your-manager http://localhost:8080/posts/talking-with-your-manager Sun, 27 Jul 2025 00:00:00 GMT "We are all fighting against the same enemy, which is a lack of revenue / productivity / engagement / time / etc. Here is something that might help us both fight this common enemy." The former approach is inherently adversarial and often leads to defensiveness and other negative emotions, while the latter is intrinsically collaborative and is more likely to be welcomed. (More of this in my article, ["Adversarial vs. Collaborative Feedback."](/posts/adversarial-vs-collaborative-feedback)) I would encourage you to offer suggestions to your manager following the collaborative approach. For instance, if you want your manager to improve on point #1 from my article above, which is, “Managers proactively let their people know where they’re spending time and effort on the unnecessary,” avoid presenting it adversarially. Don't say, "I wish you would tell me when I'm wasting time on unimportant things." Even though it's a valid request, it implies, "You've been doing it wrong, and you should follow my way because I know better," which can cause defensiveness. Instead, use a collaborative approach. Start by identifying the common problem (e.g., "Our team could generate more revenue if we focused on high-impact activities") and then suggest a solution (e.g., "Could you help me identify which of my tasks are unnecessary and what I should prioritize?"). Make sure not to blame the manager for past oversights (e.g., don't say, "You never told me what's important"), but rather take responsibility yourself (e.g., "I should have asked this sooner.") The key is to avoid any framing that implies a vertical relationship (where one person knows better) and instead aim for a horizontal relationship (where you are both working together against a common problem). ]]> <![CDATA[Tactics vs. Strategies]]> http://localhost:8080/posts/tactics-vs-strategies http://localhost:8080/posts/tactics-vs-strategies Tue, 10 Jun 2025 00:00:00 GMT growing beyond a certain point requires solving root-cause problems and acquiring long-term assets such as skills, relationships, and trust. Tactics that do not help with either can be used as a one-off when you are in a tight spot, but should not become your long-term strategy. Is there any “shortcut” you apply at work or in your life that helps you avoid facing a problem or overcome a lack of skills, relationships, or trust? If so, are you using it as a “rare one-off” or is it becoming your go-to strategy? In the latter case, the key is to think not whether using it once makes sense (it probably does, at least in the short term and at least from an emotional landscape point of view, otherwise you would not do it), but rather consider the long-term consequences of you using it over and over. A typical example is avoiding thinking about your problems. This is a fine tactic to get through a particularly stressful day. However, if you never face your problems, every day will soon become stressful. Once you have identified a short-term tactic that you should stop using as your go-to strategy, here are two questions that might help find a better long-term alternative. What root problem can you tolerate for the moment, but would not be able to tolerate for the rest of your life? What changes should you make, and what skills or relationships should you build to overcome it? _Note: In my book [“Winning Long-Term Games,”](/books/winning-long-term-games) I explain how people get stuck in life when they adopt effective short-term tactics as long-term strategies, and I explore in much deeper detail, with more extensive examples and more practical solutions._ ]]> <![CDATA[Circularity, and how to spot it]]> http://localhost:8080/posts/circularity http://localhost:8080/posts/circularity Sat, 07 Jun 2025 00:00:00 GMT ### Examples of circularity The best way to understand circularity is through some examples: - On the one hand, high IQ is associated with higher education and higher-paying jobs. On the other hand, school exams and job interviews contain IQ tests – not actual IQ tests, but questions that are better measures of the candidate's IQ than their job-related abilities. Hence, there is a circularity: part of the reason high-IQ people have higher-paying jobs is that we designed exams and job interviews to be IQ tests. - In finance, when a stock rises in price, it often triggers mechanisms that cause more traders to want to buy it, further increasing its price. The fact that the price is both outcome and cause is a circularity. - Have you noticed, when studying history, how exceptional it is that the good guys always win in the end? Victors write the history books, and historians belong to the winners. This creates circularity: those being measured (the winners) are the same doing the measurement. - On the 8th of January 2025, the New York Times titled "Meta [Facebook's holding company] says fact-checkers were the problem. But fact-checkers rule that false." Fact-checkers ruling on themselves creates an obvious circularity. These examples show that modernity is full of circularities: contexts in which those doing the measuring and those being measured overlap significantly, or when the act of measurement itself influences the outcome. ### Is circularity bad? Circularity is not always bad, though it often is, especially when we are unaware of it. For example, in science and truth-seeking, circularity makes correlations appear as causations. In finance, it might mislead traders into believing an asset is better or more stable than it actually is, causing them to enter trades whose risk they do not understand. Hence, the importance of detecting circularity: to prevent costly mistakes. ### Partial circularity Note that circularity is not a binary property, either fully present or fully absent; often, circularity is only partial. For example, finance is full of circularities, because current prices influence future ones, as highlighted by the fact that some traders buy a stock for the sole reason that it has risen. However, circularity is seldom total, for past prices are only one of the factors considered in evaluating an asset, and estimations of intrinsic value matter, too. ### How to detect circularity? Here are two tests useful to detect circularity: 1. Given two metrics A and B, does the act of measuring A influence B? For example, in the case of IQ and life outcomes, the question is not only whether A (one’s IQ) influences B (their life outcomes), but also whether the act of measuring A (administering tests that, directly or indirectly, measure IQ) influences B (life outcomes). In this case, it does, for school tests and job interview tests have an obvious effect on one’s access to good schools and good jobs. 2. Given a metric, who defines how it should be measured, and were they selected based on that same metric? For example, to some extent, it can be said that academicians are both those who decide who is smart and are selected for being smart, leading to at least some partial circularity in the definition of smartness. A positive answer to either of the points above indicates circularity. **Note: A reader submitted the topic of this article. If you also have questions you would like me to answer, let me know!** ]]> <![CDATA[Minimal Quality of Life]]> http://localhost:8080/posts/minimal-quality-of-life http://localhost:8080/posts/minimal-quality-of-life Tue, 27 May 2025 00:00:00 GMT **Minimal Quality of Life should be defined in achievable terms, without containing zero-sum elements.** Statements like "Everyone has the right to live in the most desirable city" or "Everyone deserves a top-tier degree" are self-defeating because these goods derive their value from scarcity. If we made degrees a human right, everyone would want a PhD to stand out. It's a game we cannot win, and hence we shouldn't play. This principle has practical implications. For example, the “higher education” component of quality of life should mean “access to high-quality learning content and affordable certification that doesn’t require physical attendance,” not “the right to attend prestigious universities.” Similarly, while everyone should have access to decent housing, no one can claim an inherent right to live in the specific locations that everyone else also wants, since this is mathematically impossible. Secondly, **Minimal Quality of Life should be defined by access to goods and services, not dollar amounts.** Consider housing as an example: if the average house costs $500,000 but the average family has only $200,000, we might assume each family needs an additional $300,000 to live well. However, simply giving everyone $300,000 wouldn't solve the underlying problem. Without building more housing, we would still have fewer dwellings than families, meaning some families would remain priced out regardless of their wealth. The real solution is to build enough housing for everyone. This means we should resist defining Minimal Quality of Life through minimum wage or wealth targets, and instead focus on whether we’re producing sufficient goods and services for everyone’s basic needs. Thirdly, **a definition of Minimal Quality of Life should remain truly minimal.** By definition, this standard won’t satisfy most people. If it did, we’d call it Good Quality of Life instead. But why focus on minimal rather than good standards? I address this question in the following section. ### Minimal vs Good Quality of Life Imagine we were to define the Minimal Hotel. Most people would agree on the requirements: a safe shelter, a clean room with a bed and linens, access to a clean bathroom, and electrical outlets. Maybe a few would disagree, but most would agree. Building a Minimal Hotel would be relatively cheap. We could easily, as a society, build a Minimal Hotel for everyone. Now, try defining a Good Hotel. Some would demand bathtubs, others sofas, still others restaurants or fitness centers. To satisfy everyone's definition of “Good,” we would need to include every amenity anyone considers necessary. The resulting hotel would be prohibitively expensive—impossible to provide for everyone without requiring more labor than society is willing to supply. This last point is critical. Since building hotels and their furniture requires human labor and society’s work capacity has limits, universal access to Good Hotels is only possible through variety: many different hotels, each excelling in some areas while being basic in others, allowing people to choose the trade-offs that matter most to them. All would still have to meet Minimal standards, obviously, but none would satisfy everyone's definition of “Good.” Hence, it makes more sense to measure Minimum Quality of Life and ensure everyone has a path to what qualifies as Good for them, rather than to measure a Good Quality of Life consisting of goods and services which not everyone is willing to make sacrifices for, or wouldn’t consider necessary if it weren’t for the fact that most people around them have it too. So, what should be included in the definition of Minimal Quality of Life? ### A definition of Minimal Quality of Life First, I would include all basic human rights, particularly non-physical ones: equal treatment under the law, free speech, occupational choice, democratic participation, and similar fundamental freedoms. Secondly, I would include all basic survival needs: shelter, food, medical care, and similar essentials. But in what quantity and quality? Since we’re defining Minimal Quality of Life rather than Good Quality of Life, I would include only what 95% of able-bodied people consider worth working for. If more than 5% of people wouldn’t willingly work to obtain something, it’s likely beyond what counts as truly minimal. Thirdly, I would include access to genuine paths for upward mobility. While some people may decline opportunities that require effort in exchange for improving their condition, society should ensure these pathways exist and are truly actionable. This means more than just providing access to, e.g., quality education, which is relatively straightforward in the internet age. The greater challenge is ensuring people are aware of these opportunities and, crucially, believe they can realistically leverage them. Fourth, I would include nothing else. The benefits of expanding this definition are outweighed by the costs of diluting our focus. Our attention is better spent ensuring the first three categories are not only fulfilled, but fulfilled well enough to meaningfully improve lives and efficiently enough so that we can fulfill them for everyone. This last point is of paramount importance. Just last week, in [an essay about Cultural Subsidies](/posts/cultural-subsidies), I wrote: “precisely because culture should be a common good, it is paramount to keep its costs low and cut unnecessary expenses [so that we can produce enough culture with our limited time and work].” Similarly, **it is precisely because guaranteeing a Minimal Quality of Life for everyone is so important that we should ensure we achieve it as cheaply and efficiently as possible, without deluding ourselves that it is possible to do it while keeping costs high and production inefficient.** Don’t get me wrong. I'm not lowering my ambitions for what society can achieve. Rather, I'm arguing that as long as we fail to guarantee these three fundamental requirements for everyone, it makes little sense to prioritize anything else. Moreover, it is precisely by fixing our failure to do so that we also fix the structural problems that would enable society to achieve more ambitious objectives. ]]> <![CDATA[Cultural subsidies]]> http://localhost:8080/posts/cultural-subsidies http://localhost:8080/posts/cultural-subsidies Mon, 26 May 2025 00:00:00 GMT The problem is not a lack of subsidies but an excess of costs. If my local opera theater managed to cut costs by 10%, tickets could be 20€ (assuming constant subsidies). If we believe that dropping the ticket prices from 40€ a piece to 20€ a piece would help fill that 26% of empty seats, a reasonable guess by all means, the breakeven ticket price could be further dropped to 16€. And if my local opera managed to cut costs by 20%, tickets could be free. That would be fantastic! There is a common argument that culture is a common good, just like education and healthcare, and therefore should be subsidized. I agree with it! But **precisely because culture should be a common good, it is paramount to keep its costs low and cut unnecessary expenses.** Too often, we see culture subsidies as a false dichotomy, where either we give enough money to theaters and museums to cover all their costs, or they will close. However, this neglects that while some costs are necessary (some employees, utilities, etc.), others are unnecessary (other employees, some frivolities, etc), and it is paramount to ensure the former are funded while the latter are defunded. It is also critical to ensure that the theater and museums are well-managed and run efficiently. **The idea that culture should be a basic good is incompatible with theaters and museums being badly managed and paying unnecessary wages.** This is particularly relevant since public subsidies are a finite quantity. The more we fund unnecessary expenses in one theater, the fewer theaters we can fund. (And the more we fund unnecessary expenses in theaters, the fewer schools and hospitals we can fund.) **It is precisely because culture is important that we should care about how efficiently subsidies are used.** Hence, we shall review the income statements and operating expenses of all recipients of public subsidies carefully and **make these subsidies conditional on the recipient removing unnecessary costs.** Unless we do that, we will get less culture _despite spending more on it._ #### Footnotes: [^1]: 40M CHF of subsidies for a total audience of 90k, which is 444 CHF per person, or 541 USD. Source: the 2021 audit of Geneva’s opera theater. [^2]: 21.7M€ of subsidies for 151k tickets sold. Source: the 2024 balance sheet, [^3]: Source: their 2023 financial statement. ]]> <![CDATA[The Power of Commitment]]> http://localhost:8080/posts/the-power-of-commitment http://localhost:8080/posts/the-power-of-commitment Mon, 26 May 2025 00:00:00 GMT We often think we should wait to commit until we find something worth committing to. But sometimes, it's the commitment itself that creates the worth. This doesn't mean you should commit blindly to anything. Rather, it means there's a limit to how good something can feel before you commit to it. Commitment is often what transforms something good into something great. This principle extends not only to the choice of one’s residence but also to romantic relationships, friendships, careers, and hobbies. **Waiting for perfection before committing can be counterproductive, because it is precisely commitment that creates the conditions for an option to become exceptional.** Related: [my article](/posts/twitter-network) on long-term commitment and its application to social media. ]]> <![CDATA[The 0.03-Second Margin]]> http://localhost:8080/posts/the-003-margin http://localhost:8080/posts/the-003-margin Tue, 20 May 2025 00:00:00 GMT <![CDATA[Managerial Capabilities Assessment]]> http://localhost:8080/posts/managerial-capabilities-assessments http://localhost:8080/posts/managerial-capabilities-assessments Sat, 17 May 2025 00:00:00 GMT But why is there so little time, and why do things take so long? Often, the reason is insufficient Managerial Capabilities, and improving them might actually be the fastest way to help achieve other strategic goals. ### How do you measure Managerial Capabilities? It is paramount to avoid using methods that scale, such as surveys or large data collection, for they are unreliable and suffer from the same blind spots you’re trying to uncover. For example, a common reason some team meetings are run poorly is that the people involved do not know what a well-run meeting looks like. In that case, a survey that asks how well meetings are run might collect positive feedback even in teams whose meetings have a large opportunity for improvement. A second reason that surveys are not very effective is that, when they surface problems, they do not reveal their root cause, with the risk that the findings are misinterpreted and the wrong solution selected. A much better tool to measure Managerial Capabilities is to run a series of short one-on-one interviews with people at all levels of the organization: a few top managers, a few middle managers, a few line supervisors, and a few line employees. Only in a one-on-one setting can a competent interviewer dig deep enough to uncover root causes and receive honest answers with full context. The main reason companies often prefer surveys to interviews is that the former are perceived as more time- and cost-efficient, and more comprehensive. But this is often not the case. Running a company-wide survey takes time to prepare, run, and collect the results, and requires the involvement of many people across the organization. It is faster and cheaper to task a single expert (such as myself) to run a few interviews with a few top managers, middle managers, supervisors, and line employees. Moreover, expert interviews can dig far deeper than a survey ever can, providing a more accurate assessment of root causes and thus becoming a more reliable base on which to take action. Another reason companies often prefer surveys is that they are the only realistic way to achieve full coverage of the whole organization. However, this is often unnecessary. It only takes a few interviews for an expert to have a good idea of the most significant problems and opportunities affecting the organization. He might not discover all the problems affecting all locations of the company, but neither can a survey, because the most important problems and opportunities require digging deeper than a superficial survey could. In addition to one-on-one interviews, a competent Managerial Capabilities Assessment Expert would also spend a bit of time attending internal meetings to see how they are run. Do progress update meetings help derisk projects? Do Monday Morning team meetings help attendees have a clearer idea of what to work on during the week? Are problems surfaced or hidden under the rug? Is everyone’s time used effectively? These are questions that can only be answered by observing such meetings directly. ### What to look for during one-on-one interviews First, a few things to keep in mind while planning the interviews: 1. **Dig deep, not broadly.** You don’t have to schedule many interviews (a solid starting point for a company of, say, a few thousand employees is 3-5 top managers, 3-5 middle managers, 3-5 supervisors, and 3-5 employees; three days should suffice), and you do not have to cover everything in each interview, but you do have to dig deep. Uncovering a single problem and understanding exactly what is going on is much better than uncovering ten problems superficially – the latter gives a larger impression of progress, but the change initiatives it sparks are less likely to be effective, whereas solving a single problem well often leads to many indirect improvements. 2. **Discuss what does happen instead of what should happen.** For example, do not discuss processes but what people do; and do not discuss theoretical organizational structure, but instead discuss actual interactions. 3. **Focus on habits over people.** Of course, during the interviews, it is important to uncover particularly dysfunctional people in the organization who should be terminated or particularly talented ones who should be given more responsibilities. That said, an excessive focus on individuals has limited effectiveness beyond these low-hanging fruits. Instead, a competent interviewer should surface not only people-related problems and opportunities, but also culture-related and process-related ones (and the latter should relate not only to product-related processes but also to how meetings are run and people are managed). 4. **The point of these interviews is not to catch anyone wrong but rather to uncover structural problems and opportunities.** The interviewer’s attitude should be collaborative instead of adversarial to reflect that. 5. **During each interview, indirectly collect information about the layers above and below the interviewee.** Do not ask about their boss directly, or at least do not rely on their answers in this regard, for they might be tainted by social considerations. Instead, use the interviewee’s answers about their own job to infer whether their manager provides them with enough clarity and whether they provide enough clarity to their subordinates (if applicable). That said, here is a non-exhaustive list of some of the things a competent Managerial Capabilities Assessment Expert would look for during one-on-one interviews with managers, supervisors, and line employees. - Can they explain what doing their job well looks like? (Depending on their role, probe whether they are able to explain this for their own job and/or their subordinates.) - Can they explain who is impacted by them doing their job well, and do their eyes light up as they do so? - What tasks within their team seem to take unnecessarily long? - What does their boss not know that they should know about? - What do they wish their team understood that they do not understand yet? - When there is more to do than time to do it, how do they prioritize? - What was the last problem they escalated, and what happened as a result? - If they were in charge of their department, what changes would they make? Of course, this list is non-exhaustive, and the interviewer should adapt what they ask on the fly based on the information uncovered so far and following the principle of “dig deep, not broad.” ### Experience, experience, experience It would seem that running the one-on-one interviews only requires good interviewing skills. However, an interviewer who is competent at interviewing but lacks direct management experience might not be able to spot mediocrity. For example, it is easy to notice a complete lack of clarity, but is the interviewer able to assess when a response merely indicates superficial clarity or a deep one? Similarly, is the interviewer able to assess whether Monday Morning meetings are run not merely well but also greatly (and do they know what it would take to transition to the latter?) Hence, it is important to ensure that the interviewers have a good benchmark for what great management looks like and are able to dig beyond the surface. ### Managerial Capabilities Assessment Outcomes The assessment will likely reveal a few low-hanging problems and opportunities that can be addressed relatively easily and usually at the organizational level (a few people to replace, a few people to shuffle around, and a few atomic changes to existing processes and policies). Everything else should be acted upon under the following assumptions: - **Change will be elastic unless deliberately designed to be plastic:** the former reverts to its previous form as soon as pressure is released, and the latter lasts through time. It’s of paramount importance that any initiative is implemented in a way that will result in plastic change. - **Plastic change requires everyone’s direct supervisor to be committed to it.** It doesn’t matter how well you roll out a change initiative; employees do not spend effort on what their supervisor doesn’t care about. - **Plastic change requires critical mass.** You can remind everyone once a month that they should do something differently, and they might not change during their whole career; or you can remind them a few times a day, and they will change in a week. - **Plastic change requires immediate feedback that efforts aren’t going to waste.** If you ask people to do things differently, but when they do, they do not receive immediate feedback that their supervisor noticed, they will wonder whether their efforts are going to waste, and will soon revert to their original behavior. Note: immediate means immediate; not “at the end of the week.” - **Critical mass requires shrinking the scope of change so that managers and supervisors have the bandwidth to follow up on it at least daily** (and can do so in a way that’s effective and not merely superficial). Of course, there are exceptions, but in general, you will do better assuming the five points above are always relevant rather than believing you might get away with ignoring them. ### Conclusions This was a brief overview of how I conduct Managerial Capabilities Assessments and some of the lessons I’ve learned along the way. For any enquiries, you can get in touch with me at [Luca-Dellanna.com/contact](/contact). ]]> <![CDATA[Assessing Communication Quality and Stakeholder Performance]]> http://localhost:8080/posts/assessing-communication-quality http://localhost:8080/posts/assessing-communication-quality Sat, 19 Apr 2025 00:00:00 GMT whether people focus on "ticking the box" or on "creating actual progress and derisking the project." Examples of the former include progress updates that are a recap of what happened last week instead of being about possible obstacles and how to address them, problems that are raised and noted down instead of addressed, decisions that are routinely postponed to gather more information, focus on not hurting feelings instead of being upfront, etc. Also, note whether the attitude is collaborative (“We’re all working together against the common enemy of delays, overbudgets, and low quality”) or adversarial (“We each have our agenda and they are negative-sum”). Of course, there are other ways to track the quality of communication. Still, they are all either inferior to direct observations or complementary, and I would never use them instead of direct observations. That said, here is a short overview of them: - Survey participants on the quality of meetings and communication. I profoundly dislike surveys because they are extremely superficial, and when they do surface problems, they seldom surface them with enough context to make them easily actionable. - Measure response time metrics and decision velocity. This is not a terrible idea, but it does not substitute direct observations, and if you do direct observations, you will probably not need this. - Track the rework ratio and change order frequency. In theory, this is good to track. In practice, it might invite hiding problems, and in general, it only surfaces problems once it’s too late. - Track the ratio of issues resolved vs. escalated. Same as above. - Using AI to analyze communication. In theory, it is good, but in practice, it only analyzes a subset of communication (the written or recorded part) and misses the rest. A better use of AI is to make sure people do not miss important information. ### Assessing Stakeholder Performance First of all, if you have never heard of the Pyramid of Risk, read this. A fundamental principle is that you should track the performance of stakeholders against all levels of the Pyramid, not only the top one, and you should make sure that good performance on one level is not sufficient; a good performance at all levels is required, otherwise things will turn sour soon. What should the levels of the pyramid represent? It depends on the project and stakeholder at hand, but as an indicative example, the top (incidents) could be no defect and delay in the deliverables, the middle (near misses) could be just about discussing small problems and ensuring they are addressed in such ways that they won’t happen again, and the bottom (behaviors) could be good communication quality, etc. It is important that you track both outcomes (lagging indicators) and behaviors (leading indicators). Secondly, when you assess stakeholder performance during a project (not at the end), the most important question you want to discuss is, “If things continue like this, what does the trajectory for this project lead to?” It is imperative you discuss this collaboratively and not adversarially. Thirdly, you must dig deep (and ensure that your connections in the stakeholder groups do the same). By this, I mean ensuring that: - Conversations about progress and possible obstacles do not stay generic and superficial but instead are concrete and specific. - People routinely check the situation first-hand instead of relying on second or third-hand information, such as reports. If you don’t dig deep, not only will you only discover problems until damage is done (and when discussing them will get everyone on the defensive), but you (and everyone else involved) will also miss so many opportunities for improvement. ### Addressing Discomfort I know well that everything proposed here requires going out of the comfort zone. My experience tells me it’s worth it, but that doesn’t make it any easier, of course. So, here are some considerations that have helped me in this regard in the past: Direct observations can feel like intrusions – but only if you start pointing out mistakes publicly during the observations themselves. You can prevent them from feeling like intrusions by doing the following: observe not with an attitude of “cathing mistakes” but of “understanding what’s going on and why people do what they do;” genuinely look more for best practices to reproduce rather than mistakes to correct (though, of course, both will surface); always discuss problems collaboratively and not adversarially; do not make problems about the other person (“you’re doing this and it’s wrong”) and instead refer to your experience in a way that makes the problem common (“in other projects, we had this problem, and to prevent this from happening again, let’s do this…”). If you fear “looking like the bad guy,” present all the initiatives discussed in this project not like “I, Luca, believe we should do this” (which sounds arbitrary, self-serving, and unnecessary) but rather as “Projects like this are frequently a pain for everyone because of this and that problem, and one way to avoid them is doing X and Y.” Everything you do shall be done not to serve you or the project owner but to serve the project. Half of the discomfort stems from the idea that “there is an easier way” or “you can get away with direct observations and hard conversations.” Experience teaches you mostly can’t. Discussing problems is easy when done early, before people sink their efforts and ego into something; when done late, it inherently breeds defensiveness and blame-throwing. The more you want to avoid the latter, the earlier problems should be surfaced. When dealing with multi-stakeholder projects, discomfort is inevitable. You can just decide whether hard conversations will take place before or after problems happen. I hope this helps. ]]> <![CDATA[The problem with LASIK patient satisfaction surveys]]> http://localhost:8080/posts/lasik http://localhost:8080/posts/lasik Mon, 14 Apr 2025 00:00:00 GMT at least 62% of the patients reporting that the operation created persistent eye pain still rated the operation as satisfactory. On the one hand, this indicates that patients consider the pros of LASIK to be significant, but on the other hand, it also indicates that high patient satisfaction scores cannot be taken as evidence of low risk. - In another study,[^3] **63% of patients who developed nighttime vision problems six months after the operation would still recommend the operation to friends.** Again, an indication that high patient satisfaction scores cannot be taken as evidence of low risk. - In a third study,[^4] about **45% of patients reported new visual symptoms after surgery. Despite that, the reported dissatisfaction rates were only about 1.5%.** Again, this is an indication that we shouldn’t interpret low dissatisfaction rates as evidence of low risk. I could go on, but hopefully, you get the point. **LASIK is way riskier than reported satisfaction rates imply, and anyone claiming that high satisfaction rates are evidence of low risk is a dangerous charlatan and should be called out as such.** #### Footnotes: [^1]: As an anecdote, I googled "top LASIK doctors in my town", and five out of the top ten pictures of named doctors had glasses. This is of course just an anecdote, not a solid study, but you understand why it would raise my suspicion. [^2]: Betz J, Behrens H, Harkness BM, Stutzman R, Chamberlain W, Blanco MP, Hegarty DM, Aicher SA, Galor A. Ocular Pain after Refractive Surgery: Interim Analysis of Frequency and Risk Factors. Ophthalmology. 2023 Jul;130(7):692-701. doi: 10.1016/j.ophtha.2023.02.016. Epub 2023 Feb 19. PMID: 36809816; PMCID: PMC10293021. [^3]: Bamashmus MA, Hubaish K, Alawad M, Alakhlee H. Functional outcome and patient satisfaction after laser in situ keratomileusis for correction of myopia and myopic astigmatism. Middle East Afr J Ophthalmol. 2015 Jan-Mar;22(1):108-14. doi: 10.4103/0974-9233.148359. PMID: 25624684; PMCID: PMC4302464. [^4]: Eydelman M, Hilmantel G, Tarver ME, Hofmeister EM, May J, Hammel K, Hays RD, Ferris F 3rd. Symptoms and Satisfaction of Patients in the Patient-Reported Outcomes With Laser In Situ Keratomileusis (PROWL) Studies. JAMA Ophthalmol. 2017 Jan 1;135(1):13-22. doi: 10.1001/jamaophthalmol.2016.4587. PMID: 27893066. ]]> <![CDATA[How to kill a country's economy]]> http://localhost:8080/posts/how-to-kill-a-countrys-economy http://localhost:8080/posts/how-to-kill-a-countrys-economy Thu, 10 Apr 2025 00:00:00 GMT ]]> <![CDATA[Three mistakes that derail AI adoption (and how to fix them)]]> http://localhost:8080/posts/ai-adoption http://localhost:8080/posts/ai-adoption Tue, 04 Mar 2025 00:00:00 GMT As a rule of thumb, if your training material could be given to multiple roles within your organization, it's not relevant enough and won't drive the adoption of the new tool. ### 2) Get supervisor buy-in and train them on training others People won’t change their ways just because their CEO asked them in a corporate-wide email. They need to see that their direct supervisor is actively committed to it, too. If their supervisor doesn't use AI, they won't use it either. If their supervisor doesn't care daily about whether they use AI, they won't care either. Supervisors are the lynchpin of any tool or process adoption strategy. You must work closely and frequently with them to ensure they are committed and transform this commitment into visible yet genuine actions – such as using AI themselves or coaching individuals into integrating AI into one of their sub-tasks (more on this in the next point). Moreover, you must ensure they can do that in a way that is helpful and effective – as opposed to cringe or meddling. You cannot rely on them already knowing how to do that – you must train them. ### 3) Focus on subtasks There are some jobs where AI can easily and effectively help with the main task – translators, some coders, etc. What about the rest, though? For most jobs at the moment of writing, the core task cannot be delegated to AI yet – AI is not good enough, or maybe human touch is still a requirement, as in most sales jobs, or perhaps AI would be mostly ready, but we cannot tolerate errors. Either way, when the core task of a job cannot be delegated to AI, it is tempting to default to “therefore, that job should still be 100% performed by people.” However, there are alternatives. What about the non-core tasks of a job? For example, a salesperson might still want to conduct sales in person but use AI for repetitive tasks, such as administrative ones, or to prepare for a sales meeting. Or what about sub-tasks? For example, maybe you want to keep a human in the loop of a critical task, but there are parts of it that can benefit from AI. Either way, my advice would be not to treat jobs as a monolith but rather see them as an ensemble of tasks and consider whether any subpart of it can benefit from AI – and not necessarily for full automation but either partial automation or augmentation. ### What are your challenges? I fully know that there are many more obstacles than the three I listed above – such as legal issues, ROI concerns, job security fears, lack of bandwidth, and more. I’m eager to hear about your challenges or doubts in getting AI adopted across your team or company. If you are a leader, manager, or supervisor, I’d love to chat with you. You can use [this link](https://calendly.com/lucadellanna/ai-adoption-chat-with-luca) to schedule a short call. ]]> <![CDATA[Everything I published in 2024]]> http://localhost:8080/posts/everything-2024 http://localhost:8080/posts/everything-2024 Fri, 27 Dec 2024 00:00:00 GMT <![CDATA[Static thinking is idiotic policy making]]> http://localhost:8080/posts/static-thinking-idiotic-policy-making http://localhost:8080/posts/static-thinking-idiotic-policy-making Sat, 14 Dec 2024 00:00:00 GMT A better alternative to thinking in terms of dollars is to think in terms of goods and services. Instead of asking ourselves how much a home should cost, let's ask ourselves how we can ensure we build enough homes for everyone. Instead of asking ourselves how much someone should earn to live a comfortable life, we should ask ourselves how we can produce enough goods and services so that everyone can have one (instead of having to bid money to access limited supplies). If we started doing this, it would become evident which policies lead to poverty (those that disrupt the supply of goods and services) and which lead to prosperity (those that facilitate it). Don't get me wrong. I am not naively assuming that wants are immutable. For example, even if we build enough homes to house everyone, people will complain that they cannot get a sea view. Nor am I saying we shouldn't think in terms of dollars – currencies are extremely valuable tools to facilitate trade, allocate scarce resources, and keep opportunity costs low. What I'm saying is that a large portion of the current public policy debate lacks a basic understanding of how the economy works – and with respect to that part of the discussion, forgetting about dollars and instead reasoning in terms of goods and services would produce much smarter policy-making. ]]> <![CDATA[The Chess Paradox]]> http://localhost:8080/posts/chess-paradox http://localhost:8080/posts/chess-paradox Sun, 01 Dec 2024 00:00:00 GMT It’s important to remember this. Underestimating the importance of luck is a common reason for bad decisions. Use strategies that work even if you have bad luck. ### Imperfect Execution That said, even if chess players had perfect information, there would still be a luck element. People do not always perform at their best. Even the strongest chess player might fall sick or make a blunder, and the eventuality of that happening is at least partially due to luck. Consider the following example, in which the strongest chess player plays two chess games against the second-strongest player. Both players play one game in full health and one game while sick. If they fall sick on the same day, the strongest player is likely to win both games, whereas if they fall sick on different days, the weakest player has a chance of winning the game on the day he is healthy, and his adversary is sick. In this example, whether the strongest player wins one or two games is entirely due to luck – even though, of course, were he to win both games, he would 100% deserve the victory, and it would be right to attribute it solely to his skills and hard work. Of course, it’s possible to lower the role of luck by having two players play multiple games against each other. That’s what happens in chess championships, and this is the takeaway of this post. **Even in contexts overtly about skill, luck plays a role, too. Therefore, you must manage its impact,** for example, by leveraging the law of large numbers and guaranteeing you can make a large number of attempts. ### Skill and Luck As the Chess Paradox demonstrated, **skill being of paramount importance does not imply that luck might not be important, either.** This is important because a common reason smart people fail is that they become so focused on the importance of skill and doing things right that they forget that even if they do everything right, they might still fail. **Your strategy should allow for the possibility that you do everything right and still fail. Therefore, it should contain fail-safes such as having a plan B or ensuring that even if you lose this time, you can try again in the future.** ### Case study: sports It is obvious that basketball is a game of skill. And it’s evident that the best basketball players – think about Michael Jordan, LeBron James, etc. – won because of their extreme talent and hard work. Yet, there is a reason why the NBA finals are played in a best-of-seven format instead of in a single game. It’s because luck still matters, and playing multiple games reduces its influence. You should consider applying a similar approach to your ventures. Reduce the impact of luck by making multiple bets – the more you make, the more the Law of Large Numbers will apply to you, and the more likely you will be to grab the rewards your skill would allow. ]]> <![CDATA[The Maintenance Paradox]]> http://localhost:8080/posts/maintenance-paradox http://localhost:8080/posts/maintenance-paradox Sun, 01 Dec 2024 00:00:00 GMT Plenty of activities share this characteristic: working out, spending time with your loved ones, training employees, building a solid organizational culture, and managing risks are all activities that seem unnecessary in the short term yet are indispensable in the long term. The root cause of the Maintenance Paradox is the deep-rooted belief that if we get the most out of each day, we will also make the most of the year. But it just doesn't work this way. Some actions, such as performing maintenance, have benefits that are only visible in the long term and, thus, are unaccounted for by short-term evaluations. ### Example: Why Managers Plateau The Maintenance Paradox is a common root cause of why brilliant managers sometimes fail to have a brilliant career. Managers constantly face high demands for their team’s output. So, it always feels like there’s no time for training. But unless they find the time to train their people, the situation will not improve. It actually gets worse over time as the work to be done outgrows their team’s capabilities. The solution is to switch from short-term to long-term evaluations and realize that training is not only necessary but also makes the future easier, as people will be more efficient and effective. Only managers who can shift from short- to long-term evaluations reliably succeed over the long term. ![If your time horizon is short, you will be limited in what you can achieve.](/figures/maintenance-paradox/managers.png) ### Summary Improving skills, working out, resting, strengthening relationships, performing maintenance, managing risks, and taking a step back to consider the broader picture are all activities that seem like a waste of time in the short term yet are indispensable in the long term. If you use only short-term evaluations to decide how to spend your time, you will make suboptimal choices and plateau. Instead, use long-term evaluations. This doesn’t mean you should never take any short-term action; it means to take a mix of short- and long-term actions as optimal to sustain success over the long term. ]]> <![CDATA[Bad, good, great]]> http://localhost:8080/posts/bad-good-great http://localhost:8080/posts/bad-good-great Sat, 30 Nov 2024 00:00:00 GMT it's not just about raising the bar (for example, saying: $300k in sales is bad, $500k is good, and $750k is great). It's about painting a visual representation of what a great outcome looks like. For example, imagine you tell your delegee, "If you deliver the presentation and the audience doesn't clap, it's bad; if they like it, it's good; and if they go tell their friends and colleagues, it's great." That shifts the focus from merely preparing a few slides and rehearsing them to deliberately thinking about what else must be done to make it memorable. ### The two takeaways **The first takeaway** is that merely highlighting what's enough and what's not enough is likely to lead to just okay outcomes, whereas highlighting the difference between good and great is more likely to lead to great outcomes. You can use this framework not only while delegating but also while communicating and writing job descriptions. **The second takeaway** is that this framework, "bad / good / great," is a fantastic tool for upskilling. You can dramatically improve someone's skills by showing them what great looks like (as opposed to just demanding improvement). In fact, it's such a great framework that I recently developed a workshop centered around that. It is centered around the foundational skills of people management (delegation, communication, feedback, hiring, performance management, etc.). For each skill, we will see the difference between how to do it well and at a great level. It's useful for improving your managerial skills and, more importantly, for obtaining the tools and examples to upgrade those of your team. It's a 2-hour group workshop held over Zoom. I keep the classes small, at a maximum of 12 people, and I teach it personally (no bait-and-switch with inexperienced facilitators). I will hold it on January the 29th, 2025. If you cannot make it but are still interested, let me know. I also organize private workshops. Read more about the workshop [here](/courses/bad-good-and-great-management). ]]> <![CDATA[Five thoughts on management]]> http://localhost:8080/posts/five-thoughts-on-management http://localhost:8080/posts/five-thoughts-on-management Thu, 28 Nov 2024 00:00:00 GMT Lack of clarity shows up as paralysis, busywork, demotivation, and other forms of friction ### 2. People believe you up to your consistency. If you delegate a task saying it's important but then don't check whether it got done, people won't believe it was important, and won't believe you the next time you say something is important. ### 3. People listen to you up to your helpfulness. If your feedback isn't helpful, people won't listen to it. If working on the task you delegate isn't ultimately useful for your employees, guess what, they won't work for you (beyond the bare minimum to avoid getting fired) ### 4. People are motivated by the rewards they experienced Financial incentives motivate more effectively people who have already experienced the pleasure of receiving a bonus. Start with short-term achievable targets to build trust that pursuing targets is worth the effort. ### 5. What gets discussed, and not merely measured, gets improved. Collecting metrics won't improve a process unless those metrics are discussed in a way that makes it clear that people won't be let off the hook until they improve (see #2: people believe you up to your consistency) ]]> <![CDATA[Hobby Mode]]> http://localhost:8080/posts/hobby-mode http://localhost:8080/posts/hobby-mode Tue, 26 Nov 2024 00:00:00 GMT There's nothing inherently wrong with this – as long as we correctly categorize it as hobby time and not work time. ### The importance of categorizing Hobby Mode as such I am not alone in this tendency to spend part of my work hours in hobby mode. For example, I work with an entrepreneur who has the tendency to get nerd-sniped and spends a significant portion of his work hours developing tools for his team, even though it’s not the best use of his time. We solved this by acknowledging that the time he spends doing that should be categorized as “hobby time” and boxed into some specific slot of his schedule (say, “Friday afternoons are for hobby model.”) That allows him to spend the rest of his week fully focused on supporting his team. If one truly cared about maximizing effectiveness, they would minimize hobby mode. However, life is about more than just effectiveness. Hence, it’s okay to spend some time in hobby mode. The key is to label that correctly and to be deliberate about it. ### Recognizing Hobby Mode So, how do you recognize hobby mode? It’s tricky because it looks like work and arguably has at least some work-related value. (Here, by “hobby mode,” I do not mean time spent in hobbies unrelated to work – say, gardening – but time spent doing work-related tasks that feel like hobbies.) The defining characteristic of hobby mode is a high opportunity cost. Your time is doing something of value, but it would create much more value if you spent it doing something else – or if you did that task more efficiently, only doing the necessary and avoiding the superfluous. Again, I’m not saying by any means that you should maximize work mode. However, you probably need to spend at least some of your time in work mode (depending on your desired lifestyle and other ambitions), hence the need to correctly label hobby mode as such and be deliberate about how long you spend in it. ### Hobby Mode and Innovation A common counterpoint is that hobby mode helps with innovation. This is true, but that’s a good reason not to fully avoid hobby mode – not to engage with it without bounds. Again, the key is to be deliberate about the balance between work mode and hobby mode. It’s a bit like the difference between eating and overeating. The former is healthy and enjoyable, whereas the latter is unhealthy and, frankly, not that enjoyable either. The key is not to avoid eating, obviously, but to be deliberate about it. ]]> <![CDATA[Selection Effects]]> http://localhost:8080/posts/selection-effects http://localhost:8080/posts/selection-effects Thu, 21 Nov 2024 00:00:00 GMT This means that very beautiful people will be less funny than attractive – even though, and this is key, there is no relationship between the two variables, and it's all a selection effect! ![Selecting the 10s](/figures/selection-effects/selecting-10s.png) ]]> <![CDATA[Democracies and Long-Term Games]]> http://localhost:8080/posts/democracies-and-long-term-games http://localhost:8080/posts/democracies-and-long-term-games Fri, 06 Sep 2024 00:00:00 GMT If we want to avoid large problems, we shall tackle them before they grow large enough to justify taking action. This means to regularly practice the exercises described above. And not just in the context of our country’s governance but in all aspects of our lives. Let’s imagine we executed our current strategy yet failed. What might be the most likely cause, and what can we do to prevent that? How do others with the same goal as ours fail? And what can we do about it? What does pre-failure look like? And how can we avoid that? Contrary to common belief, thinking about these questions doesn’t bring anxiety. Instead, it is precisely what enables us to sleep well (because we know we took care of the most significant risks). ### Notes This essay was written as a chapter of my 2024 book “Winning Long-Term Games,” but got cut out at the editing stage. You can get the full book and audiobook [here](/books/winning-long-term-games). ]]> <![CDATA[How to get feedback from your team]]> http://localhost:8080/posts/how-to-get-feedback http://localhost:8080/posts/how-to-get-feedback Wed, 04 Sep 2024 00:00:00 GMT Conversely, if you ask questions such as, "What is one thing I could do better?", your interlocutor will only have to think about the suggestion, not about whether to make a suggestion at all. ### 3. Don't be defensive and make them feel listened to Obviously, if you react defensively to feedback, people will stop giving it to you. However, this does not mean that you should accept all feedback. Sometimes, you have a good reason for doing something a certain way. The key is always to take your interlocutor seriously. Never question how a behavior of yours made them feel. Instead, acknowledge how you made them feel, and then explain transparently what will happen in the future. ### Summary Here is a brief overview of the three steps: ![How to get feedback](/figures/how-to-get-feedback-from-your-team.png) ]]> <![CDATA[Coaches should be experts]]> http://localhost:8080/posts/coaching http://localhost:8080/posts/coaching Mon, 02 Sep 2024 00:00:00 GMT There's nothing better than asking the coachee what they would do in a situation and providing them with some immediate feedback. It allows them to practice in a safe space and discover possible mistakes or misunderstandings without having to go through the effort and pain required by failing in the real world. ### Interested in working with me? I usually work with business leaders, managers, and entrepreneurs who want to grow their careers and organizations and believe that their bottleneck is not motivation but rather a lack of skills or context. With my usual coaching clients, we meet over Zoom once every week or every other week for 45 minutes, for 3-6 months. For some clients with specific needs, we only meet for 2-3 sessions. [Email me](/contact) for enquiries, stating your current position and career goals. ]]> <![CDATA[Values, Time Horizons, and Social Technologies]]> http://localhost:8080/posts/values-time-horizons-and-social-technologies http://localhost:8080/posts/values-time-horizons-and-social-technologies Mon, 02 Sep 2024 00:00:00 GMT Because all Values come with short-term costs and long-term benefits, they only make sense over long enough time horizons. Hence, only people with long enough time horizons practice them consistently. Of course, there are exceptions to the above: habits, peer pressure, and fear of punishment. Still, the point remains: time horizons have a massive influence on the practice of Values, for the longer the time horizon, the more sense it makes to practice the Value. ### Arbitrarily-shortened time horizons A common mistake in investing is to arbitrarily shorten your time horizon. For example, you might have 50 years to live, yet say that your time horizon is “5 to 7 years.” This is bad because it makes you take more risks than optimal (because they _seem_ to have less time to manifest) and will make you underestimate compounding properties (for the same reason). I see a similar mistake with “digital nomads” touring towns one after another, looking for one to commit to settling to. The problem is that if you think you’ll only stay in a town for three months, you won’t do things like building friendships with local communities that are key to enjoying a town. Arbitrarily shortening one’s time horizon might get in the way of exhibiting the Values that make life great. Similarly, we often shoot ourselves in the foot when we wait until a romantic relationship is great to commit, whereas some level of commitment would be necessary for the Values that make a relationship great to emerge. ### Lengthening time horizons Marriage is a social technology to increase the time horizon of a commitment to make certain Values more likely to be practiced, such as Fidelity, Honest Communication, Mutual Growth, etc. Similarly, it makes sense to ask ourselves, how can we lengthen time horizons to make Values more worthwhile? What social technologies can we introduce for this purpose? ]]> <![CDATA[What it means to be an adult]]> http://localhost:8080/posts/adulthood http://localhost:8080/posts/adulthood Tue, 27 Aug 2024 00:00:00 GMT Yesterday, I was asked about what it means to be an adult. I answered that to be an adult means to take responsibility and have learned how to commit. Taking responsibility means acknowledging that one’s actions (or lack thereof) greatly impact one’s self (and others). And learning how to commit means acknowledging that committing creates more value than retaining optionality. ### But why is commitment so hard? The answer is to be found in why many people have trouble arriving to meetings on time. Punctuality is hard if we see arriving early as something to minimize. Similarly, the trick to commitment is to stop seeing optionality as something to maximize. Don't get me wrong. Optionality and selection are still important. You shouldn't rush and you shouldn't settle. But. It is also true that building anything that matters requires time, and committing to a good but not best option produces better results than committing to the best option but too late or not fully. Adulthood is precisely this. Adulthood is the realization that commitment is not something you reserve for the optimal option but how you make good options optimal. ]]> <![CDATA[Long-term risks in investing]]> http://localhost:8080/posts/long-term-risks http://localhost:8080/posts/long-term-risks Sun, 25 Aug 2024 00:00:00 GMT if we look at the aggregate of all fund managers over the long term, Bobs accumulate more wealth than Alices. Hence, Bob's strategy is better. As I use to say, the croupier is the only person at the casino with a money-making strategy. However, every day, he sees at least one player getting wildly wealthier than him. But he must resist the temptation to switch from his good strategy to the player's worse one. ### What fund managers can learn from this The hard part of a good long-term investing strategy is to avoid the feeling of falling behind when colleagues with riskier strategies get temporarily ahead – and especially to avoid that your clients feel like they're falling behind. I can help you with this. I can both reason with you on whether a long-term strategy is truly better and, perhaps more importantly, I can give you emotional stories and effective methods to convince your clients. If you are interested, do not hesitate to [contact me](/contact). --- _Note: this example is an excerpt from my bestselling book [Winning Long-Term Games](/books/winning-long-term-games)_ ]]> <![CDATA[Pulled-forward growth]]> http://localhost:8080/posts/pulled-forward-growth http://localhost:8080/posts/pulled-forward-growth Sat, 24 Aug 2024 00:00:00 GMT The belief that pulling growth forward works is downstream the belief that if we get the most out of each day, then we also get the most out of the year. But it's a false belief. Long-term growth requires spending some days doing things that bring no immediate return but are necessary to enable growth beyond a certain point. Don’t be tricked by their early success in believing you’re falling behind and must pull growth forward in order to keep up. It’s almost never worth it. ]]> <![CDATA[Culture Wars]]> http://localhost:8080/posts/culture-wars http://localhost:8080/posts/culture-wars Sun, 18 Aug 2024 00:00:00 GMT The real culture war is between people who want to win (a more prosperous country) vs people who want to win the argument (more status even if it means less prosperity). Positive-sum vs negative-sum people. ### Who won the elections? If you look within a country, it’s easy to see who won the elections. It’s either this or that party. However, if you look from outside the country, things are different. It’s possible everyone lost the elections. That’s what happens when incompetent candidates get elected. It’s also possible that everyone won the elections. Everyone’s better off when competent candidates get elected. Competence brings prosperity, and prosperity brings more rights – that’s why conservative, prosperous countries often have better social rights than progressive but mismanaged countries (though, obviously, this phenomenon requires a few years to play out). ### Who creates environmental change? If you look at a problem such as climate change with a short time horizon, it’s an awareness problem. The only way we can improve the environment this year is if everyone changes their habits. However, if you look at the same problem over longer time horizons, decades, it’s a technological problem. I have a lower environmental footprint than my father, even if he was more environmentally conscious than me (the air quality in my town is much better than it was forty years ago because factories, cars, and heating pollute so much less). Hence, the cultural war in this context isn’t who’s willing to make sacrifices for the environment vs who doesn’t, but who’s willing to improve our technology vs who doesn’t. Who wants a solution vs who wants to be the solution. ### The path forward There have always been, and there will always be, people who want to win the argument more than they want to win full stop. People who care more about status than prosperity. More about status than the environment. More about status than everything else. That’s this excessive ego that leads them to do so much evil. They might begin with good intentions, but then they repeatedly deliberately and malevolently ignore any shred of evidence that their actions might cause harm. The solution is to call them out, over and over. Every time they focus on intentions over outcomes. Every time they focus on performance over substance. Every time they focus on zero-sum over positive-sum solutions. Every time they focus on winning the argument over winning full stop. ]]> <![CDATA[What’s better, learning from Ws or Ls?]]> http://localhost:8080/posts/learning-from-losses http://localhost:8080/posts/learning-from-losses Tue, 13 Aug 2024 00:00:00 GMT a better approach is to study the Losers. Who wrote a great book yet failed to become a bestselling author? What obstacle did they fail to overcome, and how can you prevent the same mistake? For example, did they not learn how to produce a compelling book proposal? Did they not learn how to build an audience? Of course, the two approaches are not mutually exclusive. You can learn from Winners and Losers. In fact, **unless you learn from both Winners and Losers, you will not know whether you’re learning lessons you can rely upon.** **It won’t do you much good to learn a strategy used by winners if it’s also used by losers. Unless you truly understand both what makes the winners win and what makes the losers lose, you don’t know whether you will end up a member of the former or the latter.** Don’t be like them. Constantly ask yourself, how did people who pursued the same goal as me fail? And what can I do to avoid the same mistake? If you do, you will realize that winning reliably is often more complex than apparent. There’s not just one thing you must do right. There are plenty of things you must do right. You must have hard and soft skills. You must be good at your craft and at building relationships. You must work hard and manage your health. **Fail to do any of them and the rest might not matter.** As many losers discovered. And as you might learn from them. ]]> <![CDATA[Short-Term Activism]]> http://localhost:8080/posts/short-term-activism http://localhost:8080/posts/short-term-activism Fri, 21 Jun 2024 00:00:00 GMT We choose what to do next based on whether it generates progress with no regard as to whether that progress leads to a dead end. In the case of activists, engagement obtained through vandalic arts might grow their supporters today and might continue to do so tomorrow, but eventually, they will find it impossible to grow further, once they have run out of extremists to enlist and will have to begin engaging with the bulk of the population that hates vandals. Any progress obtained through short-term tactics leads to a dead end. In the case of entrepreneurs, engagement obtained through exaggerated promises might grow their customers today and tomorrow, but eventually, they will find it impossible to grow further, after people understand they shouldn’t be trusted. The tricky part is that short-term tactics work well in the short term. Often, they work even better than long-term tactics. However, growth obtained through short-term tactics eventually plateaus. It’s a dead end. A decoy that makes it harder to achieve your ultimate objective. My suggestion to activists interested more in saving the planet than in scoring virtue points is to adopt some long-term thinking. The change you seek requires societal change, which you can achieve only in two ways: by being a tyrant or by being a leader that the majority of the population wants to follow. That requires creating trust, though. And the first step is to stop taking shortcuts that destroy trust. And my suggestion to my readers is to consider what long-term assets (trust, health, relationships, etc.) you will need to achieve your long-term goals and to avoid any shortcut that consumes rather than building them. ]]> <![CDATA[Mentorship Meetings]]> http://localhost:8080/posts/mentorship-meetings http://localhost:8080/posts/mentorship-meetings Tue, 05 Mar 2024 00:00:00 GMT ### Download the checklist Here is the checklist mentioned in the video [(PDF file)](https://luca-dellanna.com/wp-content/uploads/2024/03/Mentorship-meetings.pdf). ]]> <![CDATA[How I built a Twitter network from scratch]]> http://localhost:8080/posts/twitter-network http://localhost:8080/posts/twitter-network Wed, 23 Aug 2023 00:00:00 GMT You build your first 1,000 followers by writing value-adding replies to users with a wide following. For Instagram, it's hashtags. For YouTube, search relevance. Each platform has its own feature to leverage, and some work better than others. But what's critical is that writing great content isn't sufficient when you don't have followers – you must get it in front of eyeballs too, and simply posting to your feed is not enough. In fact, a common mistake of novice creators is to focus on creating content for their existing following before they have enough followers for it to be a sufficient growth engine. Until then, your focus should be on acquiring new followers by posting not on your feed but where these potential followers are. Twitter is a particularly good choice to build a network from scratch because users can easily retweet replies – whereas in other social media, users cannot share them or are unlikely to do so. ### From 1,000 to 5,000 If replies were the dominant way I gained followers from 0 to 1k, retweets were how I got from 1,000 to 5,000. Post great content, and your followers will show it to theirs. This is where Twitter really shines. Retweets are so frictionless and central to the user experience that going viral is easy. Other platforms have repost features too, but they are less frequently used. How do you write great content? A great way is to answer the most difficult questions your audience has. Why the difficult ones? Because they are the ones that wow people into following and recommending you. ### From 5,000 to 10,000 The growth from 5k to 10k followers happened in just a couple of weeks in February 2020, thanks to a few tweets I wrote about the pandemic as a risk management expert. Writing about current events is a good way to gain followers, but beware of how you do it. Simply covering the facts like any news outlet would make you a commodity – you will gain followers, but for the wrong reason, and they won’t be relevant to you. They will grow your follower number but won’t retweet your tweets (unless on the specific topic you wrote about when they followed you) and won’t become your customers. Instead, if you do write about current events, do it from your angle of expertise and only write about events to which your expertise is relevant. You will gain fewer followers, but they will be valuable ones, and you won’t alienate your current ones with irrelevant content. ### From 10,000 to 20,000 After my first 10k followers, the dominant way I gained followers was through shows. An account with a large following would invite me to a conference or podcast and share the recording. Each time it happened, I would gain hundreds of followers overnight. I am particularly proud of never having asked any host to get invited to their show. Instead, I published tons of expert content that demonstrated I was someone worth inviting. However, if there’s something I would have done differently, it would be to have been more proactive in contacting podcast hosts and conference organizers to get more invites. That said, I wouldn’t have been invited on shows without a track record of publishing great content. For me, it was a mix of books and tweets. For you, it can also include videos and articles. But publish first and use that to get invites – the other way around won’t work unless you’re already a celebrity. ### Beyond 20,000 As I keep growing my follower count beyond 20k, I try to observe the following principles: - C users follow B users, and A users follow A users. Writing beginner tips, news, and lists ("The top 10…") will appeal to a large audience but not to extraordinary people. To get followed by exceptional people, produce content that they would find valuable, such as advanced tips and expert commentary. Since extraordinary people are extraordinarily busy, conciseness is a must. - Reward attention, don’t waste it. My followers’ attention is my most precious asset. I shouldn’t waste it by posting something low-quality, reposting old content too often, or sharing promotions too frequently. - Do not follow a fixed schedule. Writing a daily post or weekly newsletter might force you to post sub-par content just to “fill the slot.” Don’t. Only post when you have something valuable to say. - Build trust, don’t destroy it. Don’t exaggerate claims. Don’t follow fads. Don’t compromise your principles. Be the immovable rock people can trust when the sea around it is in turmoil. - Do not imitate accounts that don’t follow the four points above. They might be faster to build a large following, but it’s going to be made of low-quality accounts. It won’t contain paying customers or large accounts that can put your content in front of the eyes of thousands. - Write content for the top nodes of your network. A retweet from someone with 100k followers will get you more visibility than 100 likes from accounts with 100 followers. Ask yourself, “What kind of content would the top nodes in my network share with their followers?” Then, produce that. - Write clearly but don’t dumb it down. People value great communicators, but – unless you’re the world’s #1 communicator, you don’t want to compete on simplicity but usefulness; otherwise, you’ll lose to someone else. - Do not become a commodity. It’s easy to copy someone else’s style or content, but precisely because it’s easy, it puts you at risk of being copied. Even worse, it makes you a commodity, forcing you to compete on aspects you don’t want to compete in, such as who’s cheaper or who works the hardest. Instead, compete on expertise. Answer better questions. Answer more difficult questions. Provide more solid answers. Be more reliable. More trustworthy. More sincere. You will attract more valuable followers. - Never measure yourself against people playing a different game. If you position yourself as the expert, do not measure yourself against people positioning themselves as the beginner’s help – if you do, you will stop writing expert content. Similarly, if you look for customers, do not measure yourself against people looking for likes – if you do, you will get more likes but fewer customers. And finally, if you’re here for the long term, do not measure yourself against people here for the short term – if you do, you will take actions that destroy long-term trust in search of a quick boost. ### My mailing list In addition to my Twitter network, I also have [a mailing list](/newsletter). I suggest you do the same, for two reasons. First, social media platforms and users come and go, whereas emails stay. Second, emails are personal and guaranteed to get on your readers’ reading list. (Whether they read it depends on whether your previous emails were worth reading.) When writing to my subscribers, I follow the nine principles above. In particular, I always ask myself the following question when writing an email: “Is this email going to make my reader more or less likely to open my future emails?” For this reason, I rarely write promotional emails. If I have to tell my readers about a new product or promotion, I write it at the bottom of an email with value-adding content (not just commodity content). I milk my current subscribers less but retain them more and – crucially – increase the chances that they recommend my newsletter to their friends and colleagues (no one recommends something too salesy). ### Conclusions To build a following from scratch, identify your network’s largest nodes, write content they would share, and get it in front of them. Differentiate yourself on reliability and expertise. Build trust more than you trade it in. Note: you might also be interested in [this interview](https://luca-dellanna.com/wp-content/uploads/2023/08/Building-a-network-on-Twitter-Luca-Wenlin-2023-08-24.mp3) where Wenlin Tan asks me about how I use Twitter and my advice for others looking to build an inspiring virtual network of interesting people. Note #2: you might also be interested in my book [Winning Long-Term Games](/books/winning-long-term-games), where I discuss how to grow long-term assets such as relationships and a social media following. ]]> <![CDATA[Metapractice]]> http://localhost:8080/posts/metapractice http://localhost:8080/posts/metapractice Tue, 13 Jun 2023 00:00:00 GMT Metapractice – the ability to tweak one's practice to maximize its effectiveness – is the most underrated skill of them all. After all, if one is bad at metapractice, he will have trouble learning any other skill. Do not just practice your skill – practice your practice. ]]> <![CDATA[Why people do not listen to your feedback]]> http://localhost:8080/posts/why-people-do-not-listen-to-feedback http://localhost:8080/posts/why-people-do-not-listen-to-feedback Wed, 19 Apr 2023 00:00:00 GMT If you want your feedback to be listened to, learn to give feedback that feels helpful, and get a track record of giving more helpful feedback than unhelpful one (as determined by its recipient). ]]> <![CDATA[Lindy for Managers]]> http://localhost:8080/posts/lindy-for-managers http://localhost:8080/posts/lindy-for-managers Fri, 10 Mar 2023 00:00:00 GMT the longer an idea or technology has been around, the longer it is expected to stick around in the future. I first learned about the Lindy Effect, also called Lindy’s Law, in Taleb’s book “Antifragile.” ### Some examples of what is Lindy and what isn’t Before discussing how managers can use it, let’s see a few examples of what is Lindy and what isn’t. - The non-perishable (e.g., ideas, technologies, recipes) are Lindy. The longer a song has been on the radio, the longer we can expect it to remain on the radio. - The perishable, such as food and people, aren’t Lindy: the older they are, the sooner they are expected to perish. After all, a 90-year-old is expected to die sooner than a 40-year-old despite a higher conditional life expectancy. - Groups of people and animals, such as cults and species, do are Lindy. The longer a species has been around, the longer we expect it to still be around. - Jobs are Lindy. The longer a job has existed, the longer we can expect it to still be around. (If you disagree bringing the counterexample of farmers, consider that farmers are still around.) - Careers are partially Lindy. The longer someone has been in politics, the more we can expect them to be in politics until they retire. The latter example gives us an insight into the two mechanics that explain the Lindy Effect. ### The rationale for the Lindy Effect There are two mechanics underlying life expectancy. On the one hand, life expectancy is inversely proportional to hazard rate: the lower one’s likelihood of dying in any given year, the longer their life expectancy. We can turn this around to say that the longer something has been around, the lower its hazard rate must have been, and therefore the longer it is expected not to perish. On the other hand, some entities (the living) have a bound to their life expectancy. People seldom live above 90 years old, and the closer they get to this bound, the more their hazard rate increases. These two mechanics sum up into a general theory of life expectancy: The longer something has been around, the longer it is expected to be around, and the closer it gets to its natural bound of life expectancy (if any), the earlier it is expected to perish. ### How can managers use the Lindy Effect? Here are a few principles we can derive from Lindy’s Law that can be useful to managers: The longer a problem has been around, the longer it is expected to be around in the future. Address it now once and for all. To estimate the life expectancy of one of your products (or one of your competitors’), consider how long it’s been around and how long the assumptions upon which it survives (habits, technologies) have been around. The more areas someone has exhibited competency in, the higher the chances they will demonstrate competency in a new area. What other examples of the Lindy Effect can you spot in your job? ]]> <![CDATA[Wittgenstein's Ruler and business metrics]]> http://localhost:8080/posts/wittgensteins-ruler-and-business-metrics http://localhost:8080/posts/wittgensteins-ruler-and-business-metrics Thu, 09 Mar 2023 00:00:00 GMT This heuristic is called “[Wittgenstein’s Ruler](/posts/wittgensteins-ruler).” I like to formulate it more precisely as follows: ### Some examples of Wittgenstein’s Ruler - When prizes (such as a Nobel) are awarded too often to a non-deserving candidate, awards tell us more about the awarding committee than about the winner. - Feedback sometimes tells us more about the person giving it than about its object. - When a ranking based on objective metrics is published, the choice of the metrics used sometimes tells us more about the committee choosing them than about what the ranking is supposed to evaluate. ### How does Wittgenstein’s Ruler apply to businesses? Here are a few examples: - When you survey your customers, make sure that you choose the questions well so that the results will give you information about your customer’s preferences for your product and not about your own preference for survey questions. - When you ask for feedback, use very specific questions and have a track record of responding well to feedback. Otherwise, the feedback you receive won’t be about what it’s supposed to describe but about your relationship with the feedback provider and your mutual expectations. - When you choose which business metrics to measure, make sure you pick metrics that describe well what’s going on and don’t instead measure the amount of “gaming” your people do to achieve them or the measurements that you’re comfortably making. Personally, I believe that in many cases, it's impossible to avoid Wittgenstein's Ruler effect entirely. Instead, it is possible to mitigate it by increasing direct qualitative observation. For example, if now and then you go to your teams' workstations and observe how they work and how metrics are collected, you might get a better overview of what's really going on in your business than any metric might tell you. I'm not saying that metrics are bad. They're good. Instead, I'm saying that metrics cannot be relied upon and must always be coupled with direct qualitative observations. ]]> <![CDATA[Ten tips to become a better presenter]]> http://localhost:8080/posts/presentation-tips http://localhost:8080/posts/presentation-tips Fri, 25 Nov 2022 00:00:00 GMT <![CDATA[Superclarity]]> http://localhost:8080/posts/superclarity http://localhost:8080/posts/superclarity Sat, 05 Nov 2022 00:00:00 GMT **be super-clear: not just enough so that you can be understood but clear enough so that you cannot be misunderstood.** Superclarity ### Why aren’t managers superclear? It is often a deliberate choice. For example, some managers believe being superclear would insult their interlocutor’s intelligence. While this might be true, it’s also true that any misunderstanding from not being superclear will damage their subordinates’ trust and respect – and the latter effect is stronger. Other managers aren’t superclear because they believe they must leave space for their subordinates to be creative in coming up with a solution. But it is possible to be superclear about an objective and yet leave space for how to achieve it. Conversely, being unclear about what you want them to do might lead to paralysis or playing it excessively safe. And finally, some managers aren’t clear because they do not know what they want, at least not concretely. ### How to become superclear The theory is easy: aim to be so clear that you cannot be misunderstood – and, to avoid micromanaging – focus on the outcome you need, not on how to achieve it. The practice is harder. I have three actions that help achieve superclarity. 1. **Be concrete and make examples.** Don’t just mention abstract concepts, such as “be ethical.” Talk in terms of actions. Explain what it means to be ethical. What does an ethical person do? And an unethical one? 2. **Ask yourself, “Let’s imagine that after I finish speaking, my interlocutor has a different understanding than me. What might they have missed or misunderstood?”** Then, of course, proactively add information that will pre-empt the misunderstanding. 3. **Ask your interlocutor to rephrase what you asked them.** This will help catch omissions and misunderstandings. Asking to rephrase isn’t an insult to your interlocutor’s intelligence – after all, rephrasing and repetition are performed mainly by high-stakes and highly-professional specialists such as surgeons and airline pilots. That said, if you feel uncomfortable asking to rephrase, you can instead ask, “what do you plan to do?” and then notice if their answer is aligned with the outcome you want them to achieve. ### Conclusion As a manager, you are responsible not only for communicating values and objectives but also for your people understanding them and, crucially, for your people not misunderstanding them. This requires embracing high levels of communication standards. Be clear and concrete. Don’t aim to be so clear that you can be understood. Instead, aim to be so clear that you cannot be misunderstood. **Be superclear.** ]]> <![CDATA[The First-Order Thinking Bias]]> http://localhost:8080/posts/first-order-thinking-bias http://localhost:8080/posts/first-order-thinking-bias Sat, 18 Jun 2022 00:00:00 GMT our minds naturally favor what I call "first-order thinking" – looking at things exactly as we experience them right now, without considering how they might change over time. Think about your first day at the gym. Your muscles ache, you're exhausted, and you don't see any immediate results. First-order thinking tells you: "This is terrible. Why would I ever do this again?" But wiser, "second-order thinking" recognizes that these initial costs will decrease while the benefits will grow. Your body will adapt, exercises will become easier, and you'll start seeing real improvements. This difference between first- and second-order thinking is crucial. First-order thinking assumes the future will be just like the present – that today's costs and benefits will stay the same forever. Second-order thinking understands that things change over time, often in ways that completely transform the original equation. Why are we stuck in first-order thinking? It's actually built into our biology. Our brains evolved to be intuitive rather than analytical. While humans can think analytically better than any other species, it takes real effort. It's mentally taxing and requires sustained focus – something our brains naturally resist. This creates a trap: First-order thinking tells us something isn't worth the effort. Because we believe this, we don't bother analyzing it more deeply. Without deeper analysis, we never discover the long-term benefits. And without experiencing these benefits, our first-order thinking never changes. It's a perfect circle that keeps us stuck. This same pattern explains why we tend to focus on negative experiences. We use our immediate reactions as a proxy for future value, forgetting that this relationship often changes dramatically over time. Just as muscles need stress to grow stronger, many valuable things in life require pushing through initial discomfort to reach lasting benefits. Understanding this pattern gives us a powerful tool: when you notice yourself dismissing something based on immediate results, pause and consider how the situation might evolve over time. Sometimes the best opportunities are hidden behind challenging beginnings. Of course, the doing the above won't be sufficient to help you change behavior; but it will help to know that our instincts are not reliable in contexts dominated by second-order effects and, therefore, we should consider acting even if our first-order emotions tell us otherwise. ]]> <![CDATA[Skill, Luck, and Imitation [Did Elon Musk get lucky?]]]> http://localhost:8080/posts/musk-luck http://localhost:8080/posts/musk-luck Sat, 18 Jun 2022 00:00:00 GMT A recent comment by Nassim Taleb made me reflect on extreme performers and what we can learn from them: First of all, let me clarify that I agree both with Taleb (most of Musk’s wealth is due to luck) and with Musk’s fans (Musk is very skilled). In fact, the two are not at odds: Musk is probably both very skilled and very lucky. Skill is sufficient for becoming wealthy, but given that one is wealthy, how wealthy depends a lot on luck. It’s easy to imagine that in many parallel universes, Musk is also a very successful billionaire, but with how many billions? One, twenty, a hundred? It would still be less than half of his wealth in the current universe. The difference, which makes up for much of his present fortune, must be due to luck. Similarly, skill is sufficient for becoming rich, but becoming the richest man on Earth also requires a lot of luck. Let me justify this statement with a thought experiment. Imagine taking a thousand entrepreneurs, all with the same starting conditions (cash, connections, etc.) and only differing in skill. Then, let’s observe how their wealth evolves in a few decades. As expected, we would notice that, in general, the more skilled a person, the wealthier they became. And yet, if you took the wealthiest one amongst them, the chances that he is also the most skilled one are slim. It’s more likely that the richest person is almost as skilled as the most skilled one but considerably luckier. My point is not about whether Musk deserves his wealth – I couldn't care less. Instead, my point is about reproducibility. We often wish to imitate the person with the highest score assuming they are also the most skilled, whereas the most skilled person is more likely to be found among those with a high-but-not-highest score. Note that I wrote “more likely.” There may be someone so skilled that he ends up with the highest score. But he must be that much more skilled than everyone else. Otherwise, the chances are that one of the “great-but-not-best” participants will get enough luckier than the best one to overcome the difference in skill. That said, my point is not “don’t imitate the best one” or “imitate the second-best one.” Instead, it’s “be critical about whose strategies you aim to imitate and why.” Ask yourself, how reproducible is that strategy? In ten parallel universes, in how many do they end up as successful as in this one, and in how many do they end up bankrupt or in jail? Are you okay with not only the outcome in the current universe but also with the full range of outcomes across universes? Moreover, consider another point Taleb made. "You get to the tail by increasing the variance (or the scale) rather than raising the expectation." In other words, to get extreme outcomes, you must reduce average outcomes. Do you really want to be the one with the highest score? It will come at a cost. Not just effort and opportunity costs but also risk – risk that will increase the best outcome but decrease the average outcome. Let me explain this last point with another thought experiment. Again, let’s take a thousand entrepreneurs, all with the same starting conditions (money, connections, etc.) and only differing in skill. Five hundred of them take extreme risks, whereas the other five hundred only take small risks. Let’s observe what would happen after a few years in ten parallel universes. We would notice that, in all universes, the wealthiest entrepreneur would come from the risk-taking group. However, and this is the key point, in each of the ten universes, it would be a different person! Each of these ten people would have extreme success when lucky and terrible outcomes when unlucky. In contrast, a skilled person who only takes moderate risks would have great-though-not-extreme outcomes in almost all parallel universes. Let me clarify a couple of things. First, this doesn’t mean that you shouldn’t take risks – here, we’re talking about extreme risks, bets that damage you irreparably when they go wrong. In general, taking small risks is a good strategy. Similarly, this doesn’t mean that you shouldn’t take extreme risks. Instead, it means acknowledging that competing for the first place often means using strategies that reduce your average outcome. Then, the choice of whether to aim for the first place is yours. And if you really want to become the best, I suggest restricting the scope of “the best at what.” For example, aiming to become the best singer in your town rather than the best artist in your country. Not only does a smaller number of competitors make it easier to win, but it also increases the chances that the winner is the most skilled rather than the luckiest one – in other words, it increases the possibility that a reproducible strategy can lead to victory (because there are fewer chances that someone gets so lucky to overcome the difference in skill with the most skilled person). This is advantageous because it means you can have a strategy that lets you win and has relatively high outcomes even if you’re unlucky. An addition, as of November 2022: the FTX crash and it’s CEO’s Sam Bankman-Fried (SBF) quick rise-and-crash from little to billions to nothing is another example of the principles described in this article. If you take extreme risks, you will have worse average outcomes than if you didn’t. As Nassim Taleb noted, “SBF got temporarily rich because he is both aggressive and clueless about finance.” And the fact that SBF overshadowed other more honest and conservative crypto exchange owners is another example of the principle, “if you don’t take extreme risks, even if you’re the most skilled person, you will be outperformed by someone who did.” ]]> <![CDATA[Too much micromanagement or too little management?]]> http://localhost:8080/posts/micromanagement http://localhost:8080/posts/micromanagement Sun, 21 Nov 2021 00:00:00 GMT Feedback on outcomes ("That's below standard" or "You crossed this boundary") is never micromanagement. 3. **Clarifying what matters is never micromanagement.** Let’s go through each of these rules of thumb. ### Feedback on methods Comments on how you would have done the task might or might not be micromanagement, depending on whether your comments are helpful. Here are some examples: - If you comment just to say something, you’re not being helpful. - If your comment adds work that doesn’t meaningfully improve the outcome, you’re not being helpful. - If your comment takes a few minutes but could have been said in a few seconds, you’re not being helpful. - Any comment such as “I would have done it this way, but it is okay if you do it that way” is a waste of time. - If you highlight that they did something wrong instead of focusing on how to improve the outcome, you’re not being helpful. - If your comment helps them see something they didn’t consider, you’re being helpful (but only if that something is important). ### Feedback on outcomes Feedback on outcomes is never micromanagement. People need to know if they are doing something well or badly, and they need to know it _very frequently_. There’s a reason many people hate their jobs but love playing video games: video games give them feedback on their performance very frequently, and their bosses don’t. ### Feedback on what matters Clarifying what matters is never micromanagement. In fact, it is management. It’s a core part of what your job is about, and if you do not do it, you are not doing your job. ### Summary Do not let your fear of micromanaging keep you from adequately giving your subordinates the information, direction, and feedback they need to perform at their best. ]]> <![CDATA[23 tips for a better career as an employee]]> http://localhost:8080/posts/career-tips http://localhost:8080/posts/career-tips Sat, 30 Jan 2021 00:00:00 GMT The #1 reason people get stuck in life is that they keep doing what worked for them in the past, even if it's not what's required for their future. Yes, being good with Excel might have been what got you the job, but it won’t be what will get you promoted. Similarly, there is this common misconception that once you get promoted, you can become more proactive. No. It’s being proactive that will make you promoted. ### Don’t do chronic overtime Occasional overtime is okay; needed, perhaps. Every now and then, there will be important milestones such as a product launch. You are expected to do what it takes to ensure that they are completed on time. However, doing overtime every week is a signal that you can’t manage your current job. In other words, that you are not ready for the next one. Your boss might be afraid that if he promotes you and gives you more responsibilities, your reaction would be to work even longer – eventually dropping some balls or burning out. No thanks, better give the promotion to someone else. Wait – what? Isn’t working overtime something that companies reward? No, not really. They’re glad if you work a lot, of course. But they won’t promote you just because of that. Here is what I would do instead. If your office hours end at 5 pm, stop working on your current tasks at 5 pm. Then, if you want, keep staying at the office – but work on something that would make you better at the job above yours. For example, learn a relevant skill, spend time getting to know senior colleagues, or think about new business opportunities for your company. But don’t do more of what you’re already doing. It won’t bring you forward much. Even better: if your office hours end at 5 pm, strive to finish your current tasks by 4 pm. Then, reserve the last hour for growth activities. This is hard to pull off at the beginning. However, this constraint will force you to delegate more and to drop the least important tasks, ultimately making you more effective. You'll also be more likely to leave your workplace on time and have more time for yourself and your loved ones. ### Your to-do list On your first day, your boss probably took some time to walk you through a list of tasks you have been assigned. Never make the mistake of believing that these are all you should do. If all you do is completing these tasks, career progression will be slow, if any. More importantly, getting better at them or working overtime on them is unlikely to bring you any tangible benefit. Instead, you should consider the list of tasks assigned by your boss to you as a starting point. Over time, you should discover which of these tasks should be done well because they matter, which should be done at the bare minimum level just to “check them off” your to-do list, and most importantly, which important tasks are not on your to-do list but should be. A good place to start is by asking yourself, what do your boss and your boss’ boss do that is not an exclusive prerogative of their role? For example, telling others what to do is a prerogative of their role, and you probably shouldn’t do that. But finding business opportunities and making processes more efficient are two tasks you might consider taking on. Of course, do not break any boundaries. For example, come up with ideas of potential clients to contact, but do not contact them unless you have your boss’s approval. But the point is: you get to the next job by showing that you would be good at your next job, not just the current one. ### Why you must discover by yourself what’s to be done Let me clarify a common doubt. If there is anything you should do, your boss would have told you, right? Wrong. There might have been many reasons for which your boss might not have mentioned that. For example, he didn’t know about it, or he put it on his to-do list to let you know but always forgets, or he hinted at it but not explicitly enough for you to notice. Alternatively, he might think that the task is someone else’s job. For example, he might think that finding new business opportunities is the salespeople’s job. Yes, it is, but that doesn’t matter that if you work in the back-office you cannot come up with good ideas for new opportunities. Nor it means that being able to recognize business opportunities isn’t a prerequisite for a promotion – especially if the next role is expected to be able to do so. I got myself into consulting from a back-office job by making suggestions no one asked me to make. Two more reasons your boss might not tell you what’s important that you do is that he’s afraid that you might take his job or that he doesn’t think that you’d be up for the challenge. Neither is a reason good enough for you not to try. Finally, there’s one last reason might not tell you everything that you need to do to get promoted. He knows that the role to which you aim to get promoted requires proactivity – and the best test to know whether you’d be a good fit is to see if you’re proactive enough to find out what you should do to get there. It’s by being who you want to become that you get there. ### Some good questions Consider asking the following questions to your boss: 1. What, of what I’m doing, is not adding value (and I could stop doing)? 2. What am I doing wrong? 3. What are you expecting me to do that I’m not doing yet? 4. Anything I can do to help you? Better to know the answers sooner than later. Don’t worry about the risk of being given more stuff to do. First of all, questions #1 and #2 are likely to decrease your workload instead of increasing it. Second, if your boss gives you additional tasks, you can always say, “there would be too much on my plate, what other tasks of mine should I give up?” Here are some more good questions to ask, this time to people in the position you aspire to reach over the next 2-5 years: 1. What are the requirements, explicit and implicit, to get where you are? 2. What did you do that helped you get here? 3. What did you do that, in hindsight, turned out to be a waste of time? The people to whom you ask these last three questions do not have to work at your company – look for answers inside and outside (but careful with contacting competitors, though, if policies or ethics forbid it; consider asking to people covering the position you seek in another industry or reading information from public sources, taking everything with a grain of salt). ### Asking for support It is your boss’s job to help you be effective. Whenever you face an obstacle that you cannot solve by yourself, or that would be addressed much faster if your boss intervened, ask him for support. Some examples of support you can request: - “I’ve never done X. I would need some training.” - “To complete my task, I need input from Legal, but they’re not replying to my emails. Can you give them a call so that they do not delay my project?” - “This task is way below my paygrade. Can we outsource it?” Of course, you should not ask for help on anything you can solve within 15 minutes with the help of a Google search or by asking your colleague next door. ### Lazy vs. Effective It will happen that one day, your boss gives you one task too much. It is your responsibility to tell him or her that you have too much on your plate. Many prefer not to ask because of the risk of appearing lazy. They try to handle everything – either burning themselves out or making some mistake. This is not good, neither for you nor for your company. Instead, you can ask in the following way, which will make your boss think not that you’re lazy but that you’re effective: “Boss, I have too much on my plate and cannot possibly do everything well enough and on schedule. Which of my tasks should I postpone or outsource?” It’s not lazy to ask for support. It’s lazy not to ask. ### Efficacy vs. efficiency Here’s a mistake that too many young employees make. At university, money was the constraint. Students do what they can with what they have. Later, they use the same approach at work. If they are given a task and a budget, they would rather achieve 80% of the target with 50% of the budget rather than going over budget. Sometimes, this is appropriate. Other times, it’s a mistake. In many companies, the constraint is not money but results. Your boss might be glad to increase your budget by 20% if that means you will reach your target but might be furious if you fail to achieve it – even if that meant money saved. Never compromise on results. If you are given a target and feel like you might not achieve it, raise your hand as soon as possible and ask for what you need. When you do so, make sure that you ask for support first, and only ask for a target relief if you’re denied the support. Asking for a target relief first does make you seem lazy. Asking for raises and opportunities The first rule of raises and opportunities is: don’t ask, don’t get. Sure, you might have seen someone getting a raise because it would be impolite not to do so after five years of loyal service. But the point is, how much more would she be paid now if she had asked for a raise the right way every 12-18 months? You must ask for raises and growth opportunities. ### A test for proactivity Many companies do not tell their employees about opportunities for professional growth, such as internal job openings or career programs. Their reasoning is that career opportunities and cash are limited resources. They should be given not just to those who deserve them but to those who have enough ambition to do more for the company. And what’s a better test for ambition than seeing what comes asking for more? Don't ask, don't get. I know; I also wish that I hadn’t had to ask for what I deserved. Sadly, that’s how things work in some companies. And if that’s representative of your workplace, there is little you can do other than acknowledge the situation and either ask for more, get yourself another job, or accept that nothing will change much anytime soon. ### The 3 ingredients to a raise There are three components of getting raises. 1. Know what is possible. In many industries, a raise every 12-18 months is realistic. 2. Do the work. You must show that you delivered value to the company beyond what’s expected from your current paygrade. Not that you worked harder, but that you delivered more value. 3. Ask for it. Make a well-thought case of why you deserve more. Let’s see each point one-by-one. ### Know what is possible You must know what reasonably talented people in your role and industry can expect as a career progression. It will make the next two points easier: you will be more motivated to put in the work, you will be more likely to ask for a raise, and you will have a better argument when asking for it. Common sources to know what a good career progression looks like in your industry is to consult dedicated websites such as glassdoor dot com, online communities, and hanging out with people who’ve been where you’re at. Yes, I know, if you ask an ethical professional how much he makes, he might not answer the question (because of confidentiality). But it’s fair to ask how frequently he asked for a raise and what it took for him to get it. ### Do the work Of course, you should do the work to ensure that you provide enough value so that when you ask for a raise, it’s justified. Also, unless you genuinely believe that you did the work to deserve a raise, you won’t feel confident enough while asking for it. “Do the work” means “do what it takes so that your boss is willing to ask his own boss to give you a raise and that he has a great argument for it.” (Yes, I know, your boss might have the authority to give you a raise without asking – but it doesn’t mean that he won’t have to justify it to his own boss.) ### Ask for it The best time to ask for a raise is perhaps during your yearly performance review, but you don’t have to wait for it necessarily, especially if it’s months away. Before meeting your boss, make sure you’re well prepared. List down the following: - The tangible value you brought to your boss or to the company, preferably with the financial value of your contribution (e.g., I closed the Nike project for $X, the process I improved saved $X, and so on). - How much you’re asking for a raise. - Why this is not much (because you can make their money back, because other companies offer similar progressions, etc.) Show you did the work. ### Handling objections The most common objection your boss might give you is, “we don’t have the budget.” This might or might not be true. Regardless, you should ask for clarifications on whether it is a temporary thing that will resolve soon. Unless you’re given a clear answer that explains step-by-step when funds will become available and from where, you should assume funds will always be an issue. Too many people waited for “just a few more months” for a raise that never materialized. At this point, you have three options. 1. Ask for a raise but not in cash. More vacation days, for example, but also growth opportunities (such as trainings, mentorships, etc.) or more interesting projects to work on. Alternatively, you can ask to be relieved from some of the tasks that are draining your energy – such as having to file travel expense reimbursements yourself or other menial tasks that can be outsourced to someone else. 2. Look for another job. If you are not satisfied with the career paths that your company is offering, feel free to look for another job – at another company, or at your current company but in a better-funded department, if any. 3. Work with your boss on a solution that would give the company enough funds to give you a raise. This doesn’t apply to all companies, but you can always propose to your boss something along the lines of: “let me pursue this new project / client / opportunity, and if it succeeds, I get a raise.” If you go for the last solution, make sure that you define what “if it succeeds” means in objective terms. Example: “if I bring a new client worth at least $200k” or “if I manage to increase production to X units an hour.” _(We’re about halfway through this post… If you think that it has been useful, please share it with your network!)_ ### Asking for a career opportunity This is not so different from asking for a raise. The points listed above apply here too. In addition, you should also list: - The times you demonstrated the skills that are needed for the career opportunity you’re about to ask. - The tangible benefits that your boss and company are likely to receive if you are given the opportunity. Be realistic but not conservative. ### When to ask for a raise or a career opportunity The number one rule is: inform yourself to what a reasonably fast-paced career progression looks like in your industry, then try to match this pace. You do not want to look for geniuses for whom work is the sole focus of your lives, but you do not want to look for average employees either. Look for reasonably ambitious people whose life you would be willing to exchange places with (both the good and the bad). Once you know how fast a good career progresses, try to replicate it. If, for example, this means a raise every 18 months, then ask for a raise at least every 18 months. ### Do not let too much time pass by It’s important that you do not let too much time pass for two reasons: compounding and precedents. Compounding. In most cases, raises depend on your previous salary. This means that failing to increase your salary once means that all future salaries of yours will be lower than they could have been. For example, let’s s say that Adam and Bob both got hired at the same time, for a salary of $2000 a month. After one year, Adam gets a raise up to $2200. Bob doesn’t. At the end of their second year, they both ask and receive a raise. Adam now earns $2400 whereas Bob earns $2200. This is true even if Adam’s and Bob’s output is the same! And what’s worse, it’s that this difference will keep compounding over the rest of their careers. This applies to promotions too. A fast early career might make your later career faster. Similarly, you might want to consider that grabbing job opportunities is easier when you’re in your twenties and flexible than when you’re thirty and perhaps with children. Setting a precedent. The more times you let pass without asking for a raise, the more you set a precedent that you’re okay with not receiving raises. This is bad for two reasons. First, your boss will be more likely to think that he can refuse or postpones eventual future requests of yours, for you’ll keep being a good employee anyway. Second, you will be less likely to ask for raises in the future or to do so confidently. If you want to go fast, you must keep the momentum up. Of course, the fact that a fast progression is possible doesn’t mean that you should go for it. Look inside you, what you want, and what your values are. Consider the negative impact that a promotion that requires a move might have on your social circle and family. Again, this essay is a map, and as with any map, it should only be used if it contains the path you want to take. ### Never accept an increase in responsibility without getting something in return For the same two reasons listed above, whenever it happens that your boss increases your responsibilities, always ask for something in return. It doesn’t have to be a raise or a promotion, especially if you received one not long ago. You can ask for more holidays, flexible working hours, better tools to do your job more efficiently, more internal support, trainings, certifications, and so on. You can ask to work on better projects or to be relieved from bad clients. (“Better” depends on what you look for: less stress? Bigger challenges? It’s up to you. But don’t let chances to make your time at work better.) ### Weekly updates Send your boss weekly email updates – especially if he didn’t ask for it. Keep them short. Do not waste his time. Don’t write what you did last week (do so only if you did something remarkable; you do not want to pass as needy). Instead, write about what you’ll do this upcoming week. Mention the bottlenecks you’re facing. Your boss might help and will anyway be more considerate of the obstacle you face. He might even improve your processes or give you support. Here’s a template. _Boss Name,_ Last week, Mark and I closed the Williams contract, worth $300k. This week, I will work on the Chicago project. I am still waiting for the contract review from Legal – could you please give them a call to ensure they send it by tomorrow? This would help ensure that the project continues smoothly. Thank you, _Your name_ Short, to the point, actionable, not wasting your boss’ time. ### Mentorship Some people find mentors to be of great help. I do agree, though I do not believe that you need a formal one. The alternative would be, at each stage of your career, to find a person whose advice can help you. Then, reach out to him and ask for help or advice. This will help you get where you want to be much faster than otherwise. ### Asking for help and advice The first rule of asking for help is to avoid wasting your mentor’s time. This means, when you first reach out to him or her, always ask for what you need directly. Do not ask for a call or a meeting. Ask for what you need. Meetings should be an option at his or her discretion, not the object of your request. This is because many people are willing to help you but have little time available. The less time you request for them, and the easier you make it for them to help you, the more chances you have they will do it. The second rule of asking for help is to explain what you will do with it. The more your mentor sees that what you want to do is impactful, the more he will be willing to help you. Of course, if there’s something in it for him, highlight it. But it’s okay to ask for something he or she will gain nothing out of it, as long as it’s impactful. The third rule of asking for help is to write personal emails. Cold emails (the technical name for an email sent to someone you never met) don’t have to be “cold” as in “impersonal.” The opposite: you should tailor them to their recipient, explaining why you are reaching out to him or her specifically and why they are uniquely positioned to help you. As a rule of thumb, if your cold email could have been sent to any other recipient just by changing the name, it’s spam. To summarize: make it easy and worthwhile for others to help you. If you do, you will be surprised by how many people will be willing to help you. If you don’t, you might end up believing that you’re alone and the world is a selfish place. ### Understanding metrics You must know and understand three sets of metrics: the ones that measure your performance, the ones that measure your boss’s performance, and the ones that measure your company or department’s performance. Unless you do, you will not be able to know what tasks are for. You will risk doing something useless and not knowing it. Also, you will miss a lot of opportunities to be effective. The three sets of metrics listed above can and should become one of the compasses you can use at work to decide what to do, what to prioritize, and what to explore (the other compass should be your values). Finally, there’s one more set of metrics you should familiarize yourself with: the metrics that measure performance at the next job in your career progression. You do not need to work on them right now, but you must work on the skills that will be needed – and eventually find ways to showcase them. After all, that’s one way to demonstrate you’re ready for your next job. ### The three factors that influence your career Three factors will have a disproportionate influence on your career. 1. How good you are at your job. Obviously. 2. How fast is your department / company / region / industry growing. The faster it grows, the more budget and opportunities will be available. 3. How good is your manager, and how aligned your values are. A bad manager might become your worst nightmare and significantly affect your income and stress levels, and thus your life outside of work. Too many people focus on the first one and leave the other two to chance. Don’t. The impact work has on your life is too important. Of course, this doesn’t mean “forget about your childhood friends and dreams and go to work in tech in San Francisco.” It means to know that where you work opens some doors and closes others, and that good opportunities can be found in the unsexy city if you look for them in the right industry. Make an intentional choice based on your dreams and values, and do not let it up to chance. Similarly, your boss might be the best or worse thing that happened in your career. While interviewing for a job, make a point of meeting your new boss or ask your potential future colleagues about him or her. If you find yourself with a terrible boss, find a way to get a new one (for example, by transferring to another team or company). Otherwise, he or she might wreck your career, health, and personal life. ### A recap of the most important points - Doing more of what got you here won’t get you there. - Don’t work more hours than necessary; if you want, use the extra time to do what will get you your next job, not more of the work you’re already doing. - Ask for what would make you more effective – it’s in everyone’s benefit. - Know when you should ask for raises. Prepare yourself by showing the value you delivered in $$ terms. Handle objections. - When asking for help, don’t waste your recipient’s time and make it easy for them to help you. - Know the metrics that matter to your job, your boss’, and your company. - Don’t leave to chance who you work for and who’s your boss. ### Conclusions These were a few of the things I learned too late about career in a corporate environment. Hopefully, they will be of help to you. I wrote a book from a similar point of view, called [100 Truths You Will Learn Too Late](/books/100-truths-you-will-learn-too-late). It got great reviews and helped hundreds of people. You can download a copy here. If you know anyone who could benefit from this blog post, please share it with them! Also, you might want to subscribe to [my newsletter](/newsletter) to receive future ones. ]]> <![CDATA[Just In Time]]> http://localhost:8080/posts/just-in-time http://localhost:8080/posts/just-in-time Sun, 06 Sep 2020 00:00:00 GMT The solution is to keep some stock, but make it hard to access. For example, a signature from the Operations Manager might be required to pull stock from the warehouse. This way, you have some protection against logistics disruption, and you partially cut storage costs and surface problems. ]]> <![CDATA[Bottom-Up Manifesto]]> http://localhost:8080/posts/bottom-up-manifesto http://localhost:8080/posts/bottom-up-manifesto Wed, 16 Oct 2019 00:00:00 GMT Long-lasting progress is always bottom-up, because the bottom-up becomes adopted after having being proven to work. The top-down is fragile because untested. - The flow of life is bottom-up. - The arrow of causality is bottom-up. History repeats itself because history is made by the bottom-up, not by the top-down, and the bottom-up is relative, and the relative takes place over and over again, just like we all aspire for a better life for ourselves, no matter how good our life is. Top-down practitioners consider their peers to be the ultimate judges of an idea; bottom-up practitioners consider reality to be the ultimate judge of an idea. Top-down that works is impossible. It necessitates full knowledge of the territory, and no single sensor could be tuned perfectly to make every detail of the territory legible (it would have to be calibrated to a single part of the territory). We are under the illusion that there is something top-down which can influence the world in a lasting way. Either it won’t last or it’s something which emerged the bottom-up way and we confabulated it as top-down. Examples: inventions, policies, thoughts, careers, successes. Less time should be spent on discussing top-down solutions and more on creating the conditions for bottom-up ones. ]]> <![CDATA[Fragilization]]> http://localhost:8080/posts/fragilization http://localhost:8080/posts/fragilization Wed, 15 Aug 2018 00:00:00 GMT **antifragile entities cannot become fragile (i.e., they do not lose the property to positively react to stressors in a given range). However, they can have such range reduced. I call fragilization such reduction of the range of stressors that benefits them, and antifragilization its increase.** ### The conditions for antifragility Antifragile entities have a common property: they comprise a population of sub-entities at a lower layer. Some examples: - Animals and plants are composed by cells. - Species are composed by a population of animals or plants. However, also ceramic vases are composed by a population of ceramic molecules, and yet they are fragile. Something is missing in the previous definition. Let’s try again: Antifragile entities have a common property: they comprise, at a lower level, a population of sub-entities which are subject to natural selection. Some examples: - Animals and plants are composed by cells; if subject to a stressor, the weakest cells die. - Species are composed by a population of animals or plants; if subject to a stressor, the weakest members die. Natural selection is what provides antifragility. Natural selection necessitates stressors (to kill the weak). However, some might say: “hey, also the molecules of a ceramic vase undergo natural selection: if you make a vase fall, some of the molecules that form it (or rather, some of the bonds between them) will break; moreover, it’s not random ones that break: it’s the weakest ones.” Something is missing, again. **Antifragile entities only exhibit antifragile behavior when their underlying population dies in a distributed way.** Some examples and counterexamples: - Animals are composed by cells; if subject to a stressor, the weakest cells die. If the weakest cells die in a distributed, non-local way, the exhibited behavior is antifragile (lifting weights at the gym breaks only a few muscle cells here and there, and the muscles regrows them in excess, becoming stronger). If the weakest cells die in local way, the exhibited behavior is fragile (if I lift a very heavy high, my muscle might strain and I lose the ability to use it properly for a few days; if I jump from the third floor, all the cells in a given area of my leg bone break and the bone breaks). - Species are composed by a population of animals or plants; if subject to a stressor (such as a famine), the weakest members die. If the members die in a distributed way (some members of this tribe, some children of this family), then the tribe or family grows stronger (thanks to natural selection and learning): an antifragile behavior. However, if all the members of a local population die, then, simply, that tribe is exterminated, without opportunity to adaptation (through natural selection and/or learning): a fragile behavior. ### The disappearance of antifragile responses In the previous part, I described how an entity needs four conditions to be able to express an antifragile behavior: - to be composed of a population at a lower layer, - to allow for natural selection to act, - to ensure that natural selection kills its lower layer in a distributed way, - to be able to grow their lower-layer population up to a point of redundancy. In the rest of this essay, I will describe what happens if natural selection is prevented (through the removal of skin in the game or through the introduction of “fences”) and what happens if conditions are present which might cause its population to die in a non-distributed way. In both cases, we will see, the consequence is fragility. First, let’s see how a fragile entity reacts to stressors. ![A fragile entity](/figures/dellanna-diagram-the-fragile.png) _Note: the above is called a ["Dellanna Diagram"](/posts/dellanna-diagrams)._ If the intensity of the stressor is lower than the fragility threshold, no change happens (if I let a ceramic vase fall from 1 inch, it doesn’t break). If the intensity of the stressor is higher than the fragility threshold, the fragile entity breaks because its lower-layer population incurs a localized death (if I let a ceramic vase fall from 3 feet, the molecule bonds along a weak crystal boundary break and a crack appears). (For the time being, I am applying a simplification: considering single stressors only. As engineers know, some entities exhibit a behavior called “fatigue failure”, where the effects of smaller stressors accrues over time. For the purpose of explaining the current concepts, I am only considering single stressors. Later in the essay, I will clarify the effects of distributions of stressors.) Now, let’s see how an antifragile entity reacts to stressors. ![An antifragile entity](/figures/dellanna-diagram-the-antifragile.png) If the intensity of the stressor is higher than the fragility threshold, the antifragile entity breaks because some of its lower-layer population breaks in a localized pattern (for example, if I jump from the second floor, I might break a bone). If the intensity of the stressor is between the fragility threshold and the antifragility threshold (the green area above), then antifragilization happens. For example, if I jump fifty times over a 3-feet tall obstacle, I will grow some leg muscles. Antifragilization happens because the lower-layer population dies in a distributed way (for example, I break some muscle fibers here and there: this triggers muscle regrowth. If I exercised too hard, I would broke the muscle fibers in a localized way and I would have a muscle strain: an injury, a fragile behavior). I call this behavior antifragilization because it modifies the future response of the antifragile entity to future stressors. Following the effect of the stressor in the antifragilization area, the fragility threshold moves to the right, as the entity grows stronger and able to withstand more intense stressors without breaking (the more I go to the gym, the more I can lift without straining myself). As a result, there is a bigger proportion of stressors which induce an antifragile response in my entity: it became more antifragile. Hence, antifragilization. I will describe this behavior better later on. If the intensity of the stressor is, instead, lower than the antifragility threshold (the orange area in the image above), no member of the lower-layer population dies. For example, if I lift a matchbox, the weight is too low to break any of my muscle fibers and to induce any muscle growth. If, over time, no stressor happens to be higher than the antifragility threshold, then fragilization happens. If I do not exercise for a few of months, my muscles mass will decrease, and I will lose the ability to lift heavy weights. In such case, both the antifragility threshold and the fragility one decrease (moving to the left in the image above). Because the fragility threshold lowered, the antifragile entity is more likely to express a fragile behavior (because, given a stable distribution of stressors, now a bigger proportion of them falls above such threshold). Therefore, fragilization, I will describe this behavior over the next paragraphs, just after antifragilization. ### Antifragilization ![Antifragilization](/figures/dellanna-diagram-antifragilization.png) As displayed in the image above, antifragilization (the process that happens when an antifragile entity is exposed to stressors whose intensity falls between the fragility and antifragility thresholds), consists in a distributed death, followed by an adaptation and therefore an increase in both the antifragility and the fragility threshold. The increase in the fragility threshold is bigger than the increase in the antifragility threshold, so that the range of stressors inducing an antifragile reaction is bigger, and therefore the antifragile entity becomes more likely to exhibit an antifragile reaction. ### Why does antifragilization happen? Antifragilization is the result of adaptation following natural selection. Step by step: 1. A stressor hits the antifragile entity. For example, I lift some weights. 2. The stressor is strong enough to induce natural selection: it kills the weakest members of the lower-level population. Continuing the example, lifting the weights causes some of my muscle fibers to break. 3. In antifragile entities, those weakest members are usually normally distributed; therefore, a distributed death occurs. The distributed death is well-absorbed by the entity and therefore does not cause it to permanently lose any functionality. (I will describe later what happens if, due to some external intervention, the weakest members are not distributed and therefore a localized death takes place.) Continuing the example, the muscle fibers which break are not grouped in a single location (as it would happen in case of a strain) but are distributed along all the muscle. 4. A first effect, is that since the weakest members of the population died, the average survivor is stronger. Only the strongest will reproduce, generating a second-generation population which is stronger than the first one on average. Continuing the example, it is the weaker muscle fibers that die; only the strongest fibers will be left to reproduce, so that the muscle fibers that are born will be strong, on average. A variation of this effect, applicable to populations of humans, is that of learning. Following the negative event, the members of the population that were practicing the weakest beliefs die, leaving only the members with the strongest beliefs alive and able to reproduce. If a source of water in the village is contaminated, only those members of the population who thought that boiling/alcohol are not necessary to purify the water will die, leaving those who thought that boiling/alcohol are necessary alive and able to pass their advantageous beliefs to the new generation of the population. 5. As a second effect, the intense stressor creates an expectation of more intense stressors to come, so that the need for redundancy is established, and overgrowth is triggered. Continuing the example, lifting weights creates the expectation, in my body, that it will have to lift more weights in the future; it better prepare itself for the task by growing muscles which are not only strong enough to lift the weight I just lifter, but even stronger to lift even heavier weights. This is the process behind muscle mass growth at the gym. [In what resembles a bayesian interpretation of the world, after being exposed to an intense stressor, our body intuitively modifies the distribution of future expected stressors, so that even stronger stressors, never experienced before, are deemed possible.] 6. The consequence of the two effects described before is an increased tolerance to stressors, which causes both the antifragility and the fragility thresholds to increase. ### Fragilization ![Fragilization](/figures/dellanna-diagram-fragilization.png) Fragilization is the result of the lack of stressors and, thus, of natural selection (which needs stressors to kill the weak). Step by step: 1. For a certain time (weeks, usually), an antifragile entity lacks exposure to stressors above the antifragility threshold. For example, for three months I do not make any physical exercise. 2. In that period of time, natural selection does not take place in the population at the lower layer (at least along the dimension of the lacking stressors). Continuing the example, in those three months no muscle fiber breaks because of an applied physical stressor. 3. A first effect takes place: because of the lack of natural selection, the weakest members of the lower-layer population do not die and are therefore allowed to reproduce. The second generation population has thus a lot of weak members, which means that both the antifragility threshold and the fragility ones become lower. Continuing the example, the weaker muscle cells are allowed to reproduce, and therefore the muscles contain an increasing proportion of weak cells. 4. A second effect takes place: because of the lack of stressors, there is no (apparent) need for redundancy anymore. Redundancy is useful only in bad times; in good ones, it seems unnecessary. Therefore, the antifragile entity adapts in such a way to reduce redundancy. This makes sense, because optimization is an adaptation to stability (for it assumes certain conditions, to which the entity optimizes for, to be present also in the future), whereas redundancy is an adaptation to volatility (for, if no condition is assumed stable, there is no stable condition to optimize for). The lack of redundancy makes the antifragile entity more vulnerable to failure, and therefore more fragile (the fragility threshold becomes lower). Continuing the example, because I didn’t lifted any weight for months, my body assumes I will not have to lift any weight in the future either. It then appears advantageous to reduce the number of muscle cells (to save energy, between other reasons). This makes me more likely to strain my muscles in case in the future I will suddenly have to lift an heavy weight. 5. The consequence of the two effects described before is a decreased tolerance to stressors, which causes both the antifragility and the fragility thresholds to increase. ### Antifragile entities are adaptive As described with the two processes of fragilization and antifragilization, the range of stressors which induce an antifragile reaction in an antifragile entity is not fixed. Antifragile entities are adaptive: this confers them an advantage in a changing environment. However, precisely because they are adaptive, they can also adapt to a stable environment. **Adapting to an unstable environment makes an entity “more antifragile”** (i.e., more likely to exhibit an antifragile reaction and to adapt to its environment), **and adapting to a stable environment makes an entity “less antifragile”** (i.e., less likely to exhibit a fragile reaction and to instead not adapt to its environment). Again, I must clarify: antifragile entities never lose their capacity to being adaptive (i.e., to benefit from stressors in a given range); however, they might shrink the range of stressors towards which they react in an adaptive way. Therefore, **though antifragile entities never become fragile, they might change their likelihood to react to stressors like a fragile entity.** ### The importance of natural selection As described above, lack of natural selection causes fragilization. There are mainly two processes which prevent natural selection: the removal of skin in the game and the introduction of exogenous barriers (aka “fences”). The next part of this essay will describe them. (Side note: hacks and quick fixes, which which offer apparently safer shortcuts but tend to bypass or alter natural selection, bring fragility.) ### Introduction of skin in the game First, let’s see what happens if skin in the game is increased in the population forming the lower layer of the antifragile entity. For example, if a human nation (an antifragile entity) passes a law to enforce skin in the game for its citizens (the population at the lower layer). ![Skin in the game](/figures/dellanna-diagram-skin-in-the-game.png) 1. First, the introduction of skin in the game introduces direct accountability: one is more likely to suffer for his mistakes. Since mistakes usually become evident not when things go well, but the moment they turn awry, it is usually following the application of an external stressor (such as a financial crisis) that an individual with skin in the game exits the population (for example, it goes broke or gets fired from his investment firm). The more skin in the game, the more, as a first-order effect, weak individuals are likely to fail in case of stressors; such failure is likely to be distributed (thanks to skin in the game, only the weak individuals fail, without bringing anyone down with them), therefore, it is the antifragility threshold which gets lowered. 2. Because the antifragility threshold became lower, the antifragililization range (the green area) become bigger. 3. A random stressor is now more likely to induce antifragilization (a stressor which would have hit the right side of the orange area would now hit the left side of the green area). Therefore, antifragilization is more likely to happen. 4. As a result of antifragilization, the fragility threshold increases. As a consequence, a random stressor is now more likely to hit the antifragilization area (a stressor which would have hit the red area in its left side would now hit the green area in its right side), and the entity is now more likely to exhibit an antifragility response. 5. Because of the antifragile response being solicited more often, antifragilization happens more often, and the fragility threshold now increases moving to the right and offsetting the movement it did to the left during the first step. As a result, **thanks to skin in the game, all members of the population are now safer (less likely to die because of a random stressor).** It is worth noting that the dominant effect, and the only one significant over the long term, is the one described in the fifth point above, a second-order effect. However, because policy makers usually only think terms of first-order effect, they might take wrong decisions, some that only appear to work in the short term. Now, let’s see what happens in the opposite case, the removal of skin in the game. ### Introduction of exogenous barriers (aka, “fences”) and the removal of skin in the game Humans do not like risky environments, and often introduce barriers to make them safer. However, often such barriers only increase the sense of safety. As Pasquale Cirillo describes in [“The Fence Paradox”](https://x.com/DrCirillo/status/727526381878403072), because the barriers reduce the smaller stressors, humans feel safe, they take more risks, increasing the risk of an important negative event once a stronger stressor appears. In The Fence Paradox, Dr. Cirillo tells a story of tourists taking pictures from the edge of a canyon. Since it is a risky endeavor, only a few tourists venture to the edge, and only do so with great care. One day, one of them goes a bit too far on the edge and falls down, dying. In response to this unlucky event, a fence is build on the edge of the canyon. It now looks safe to take pics from the edge, and tourists flock in. They lean on the fence to take pictures, and push each other in an attempt to take better shots. One day, a big group of tourists relays too much on the fence and they apply too much of their weight to it: the fence collapses, causing the whole group leaning on it to fall and die. After all, the fence didn’t make the canyon safer; it only made it safer to the population exposed to (or exposing itself to) smaller stressors, whereas it made it less safe in case of exposure to bigger stressors (or making the population more likely to expose themselves to bigger stressors). The image below describes the fragilization which follows the introduction of exogenous barriers (which, by causing the population to reduce their exposure to the consequence of some of their actions, is not so different from the removal of skin in the game). ![Exogenous barriers](/figures/dellanna-diagram-exogenous-barriers.png) 1. First, the introduction of exogenous barriers reduces direct accountability: one is less likely to suffer for his mistakes. As a first order effect, weak individuals are less likely to fail in case of (small) stressors; therefore, the fragility threshold increases. 2. Because the antifragility threshold became higher, the antifragililization range (the green area) becomes smaller. 3. A random stressor is now less likely to induce antifragilization (a stressor which would have hit the left side of the green area would now hit the right side of the orange area). Therefore, antifragilization is less likely to happen and fragilization is now more likely to take place. 4. As a result of fragilization, the fragility threshold decreases. As a consequence, a random stressor is now more likely to cause death (a stressor which would have hit the green area in its right side would now hit the red area in its left side), and the entity is now more likely to exhibit a fragile response. As a result, **thanks to the introduction of exogenous barriers, all members of the population are now less safe (more likely to die in a localized way in case of exposure to a strong stressor).** ### The importance of redundancy In a first-order analysis, redundancy might seem to reduce the likelihood of antifragilization, for there is a higher probability that an exterior stressor is not perceived interiorly. For example, if I am strong (i.e., I have redundant muscles for most daily activities), I will not grow muscles by lifting my not-too-heavy grocery bags, whereas if I am weak and start lifting heavy grocery bags every day, I might actually grow some muscles. However, a second-order analysis shows the opposite consequence: redundancy increases one’s willingness to expose himself to stressors, and/or his opportunities to do so. A strong person is more likely to go to the gym [footnote] and a person going to the gym is more likely to be strong; the two affirmations are not at odds[/footnote], whereas a weak person is less likely to expose itself to stressors (for example, to offer to help to move furnitures). Moreover, a stronger person (i.e., a person with redundant muscles) is more likely to survive a stronger-than-usual stressor, meaning that it is more likely to survive to benefit from the stressors. A weak person (one with no redundancy) is more likely to injure itself or to die in case of extreme physical stress, and therefore less likely to continue its exposure to the smaller, more beneficial stressors that will take place in the future. Therefore, a certain degree of redundancy is beneficial to antifragility. ### How to become antifragile? To conclude, there are four necessary conditions for antifragility, and four actions which can increase it. The 4 necessary conditions for an entity to be antifragile: 1. It has be composed of a population at a lower layer, and 2. Natural selection has to be allowed at that lower layer, and 3. Killings from natural selection have to happen in a distributed fashion, and 4. The surviving members of the population have to be able to reproduce. The 4 actions which can increase the likelihood of an antifragile response to a random stressor: 1. Ensuring skin in the game. 2. Removing exogenous barriers (aka, fences). 3. Encouraging redundancy. 4. Reducing sistemicity. It is worth noting that the 4 action points above only work as second-order effects, and that understanding them requires a dynamic representation of the world. People using static representations of the world (IYIs), and therefore unable to understand second-order effects, are likely to dismiss them as useless. _Note: this was an excerpt from my book, ["The Power of Adaptation"](/books/the-power-of-adaptation)._ ]]> <![CDATA[Ruthlessly Pareto]]> http://localhost:8080/posts/paretoing http://localhost:8080/posts/paretoing Tue, 07 Aug 2018 00:00:00 GMT If the rewards (outputs) of a given activity are nonlinear, the inputs shouldn't be either. Some examples: - If you’re reading the same percentage of pages for each book, you’re doing it wrong. (given that time is a finite resource). Some books are worth skipping after having read a few pages, whereas other books are worth reading many times. - If you are attending all classes or all meetings, you’re doing it wrong. Some are worth not attending, and some are worth preparing for before and reflecting upon later. - If you are spending the same amount of attention on each customer of yours, you’re doing it wrong. Some customers are worth following with double the attention, some are worth the minimum, and some are worth firing (you might discover that 20% of your customers generate 80% of your customer service costs). - If you are spending the same amount on Twitter each day, you’re doing it wrong. Some days, you’ll be inspired, or great conversations will be going on, and other days, you’ll be just passively scrolling. - If you are spending the same amount of hours at work every day, you’re doing it wrong. Unless you are the lowest employee at your level in a super-optimized operation, chances are that some days there will be opportunities worth spending more time on, and other days could be finished much earlier. And so on… Ruthlessly Pareto your activities. ]]> <![CDATA[Bad decisions in life arise from having optimized for the wrong metric]]> http://localhost:8080/posts/optimizing-for-the-wrong-metric http://localhost:8080/posts/optimizing-for-the-wrong-metric Mon, 17 Jul 2017 00:00:00 GMT ### Proxy optimization I once wrote: “Humans are extremely good at succeeding at their priorities, and extremely dishonest about them”. With that, I meant that we are great at acting in such a way to succeed at the metric we chose, but often choose the wrong one or do not realize we subconsciously chose another one. I call this phenomenon commitment to failure. Every time we fail at something we had the resources to succeed at, it is due to the fact that we considered an internal success to fail at it (because we were uncomfortable with the consequences of succeeding at it). In other words, we acted optimizing an internal metric (Expected Emotional Outcome) instead of an external one (success at whatever activity we were doing).[^1] Even when we choose a proxy that is initially correlated with the ultimate outcome we desire, we often fail at misunderstanding a key concept:[^2] As an example, imagine an author in the business of selling books online. He discovers that about 2% of his Twitter followers bought his book. He decides to run an advertisement to get more followers (the number of followers becomes a proxy for book sales). Such ad selects followers based on their sensitivity to advertisements, not on their propensity to buy. In other words, it tends to select people who are interested in following the author on Twitter, regardless of whether they are interested in buying his books. As a result, if before the advertisement 2% of followers would end up as customers, after the ad it it is likely on 1% will. The proxy optimization (running an ad optimized to get new followers) diluted the correlation between the proxy and the ultimate metric (the % of followers buying the book). A useful heuristic: if there is a way you could maximize the proxy without contributing to your ultimate goal, it will happen. Proxies are not chosen because they are more impactful; they are chosen because they are easier to measure or to influence. And the reason they are easier to influence, is generally because there is a population of low-hanging fruits which contribute to the proxy but not to the ultimate metric. Collecting these is a waste: they do not do anything to improve your standing towards your ultimate goal. Optimizing for a proxy often brings a dilution in quality. Proxies promote unwanted behavior by introducing unwanted rewards towards wasteful actions. It is the same thing that happens when you are not specific in rewarding someone. Let’s say that you told your son “Congratulations for having gotten an A on the test!”, without knowing that he cheated. By rewarding a proxy (the mark) instead of the real thing (the study), you would have reinforced an unwanted behavior (the cheating). ### Define success without using proxies Naval Ravikant said: “A cockroach is just as evolved as we are, just across different fitness functions.” What metric you choose to define success, i.e. your fitness function, matters. It will apply evolutionary pressure on your behaviors and personality traits and shape you, by reinforcing those traits that led to a progress on your chosen metric and by weakening those who did not. Choose your proxies wisely, don’t become a cockroach. Another reason why proxies should not be used, is because circumstances change over time. It might sound useful to choose a proxy to focus on what appears to be the most beneficial sub-objective now, but this might lead to tunnel vision or to a lack of periodic reevaluation to determine whether the current proxy is still beneficial to follow. What got you here won’t get you there[^3]. Memorizing concepts might be a useful skill to get your degree, but it is a terrible one for a successful career (other than in acting). Money is great to increase your life quality when you’re poor, but it doesn’t do much once you’re rich. **What looks a great proxy now will almost invariably be a bad proxy then.** A last reason not to use proxies: they bring noise (because they are partially decoupled from the ultimate metric which we do want to measure) and they bring cognitive dissonance (for the same reason; if we ever take an action which is good from the point of view of the proxy but not of the ultimate metric, we’ll end up asking ourselves “Why did I did it?”. Following proxies instead of the real thing is similar to drug addiction. Addicts become reactive to the cues (the proxy). Living a diverse life increases happiness because it avoids addiction and tolerance due to repetitively associating rewards to the same cues[^4]. (I am inclined to believe that humans are fundamentally good and happy; evil and unhappiness emerge when they follow unnatural proxies or when they encounter constructs or products engineered to hack metrics, such as advertisements and drugs, and become addicted to them). ### Eliminating proxies Prioritization is the art of spotting proxies and exclude them from our to do list. Proxies are fought with not-to-do lists. A good example is Warren Buffet’s 25–5 rule. Write down the 25 most important things you should do. Circle the 5 most important. Move these 5 to your to-do list, and the other 20 to your not-to-do list (these 20 tasks would be busywork, which, by the way, is a proxy for real work). Intelligence is the skill of optimizing a given metric; wisdom, the skill of choosing the right one to optimize for. Some heuristics on how to spot proxies and get rid of them: - Everything which causes addiction is a proxy. - Any metric which is not weighted by the impact it generates is a proxy. (And any metric whose correlation/impact to your ultimate goal is approximated to be static in time, is a proxy.) - If a proxy is chosen because it is easier to improve, rather than easier to measure, don’t choose it. - If a process describes what it is, rather than what it is for, it has been optimized for a proxy. - If it won’t be important in five years, it is a proxy. - If it weren’t important for people one thousand years ago, it is a proxy. - Doubt everyone whose core competency is a proxy (students whose competency is to pass exams, which are a proxy for knowledge; teachers whose competency is to get papers published, which are a proxy for teacher quality; financial advisors whose competency is looking trustworthy; the salesman who’s great at presenting a slide deck but not at actually making sales happen; speakers whose competency is to use words rather than choosing them; and so on)[^5]. These heuristics won’t be always true, but you’ll be better off by acting like if they did (that’s the point of heuristics). _This was an excerpt from my book [“100 Truths You Will Learn Too Late.”](/books/100-truths-you-will-learn-too-late)._ _Many thanks to Damian Ocean for providing insights to the first drafts of this post, which appeared on Twitter._ #### Notes [^1]: Alexander Wolfe noted that “What people who don’t risk don’t understand is that it is better to risk and fail at something that resonates deep within than to do nothing”. To which I replied: “That’s the difference between external failure (proxy) and internal success (actual metric to consider).” [^2]: This is not (only) a repetition of Goodhart’s law. Goodhart focused on people gaming the system once a regulator introduced a policy. My heuristic is more general and works in single-player scenarios too. It is because the easiest way to improve an unweighted metric is often to include items whose weight towards the ultimate goal is lower than the past (e.g.: in the example in the text, the easiest way to improve the number of followers is to attract those who won’t buy anything). [^3]: As far as I know, Marshall Goldsmith made the expression famous, in his homonymous book. [^4]: Living a diverse life means acknowledging that because of the problems of addiction and tolerance, pursuing a single proxy will yield diminishing returns over time and lead to dissatisfaction. [^5]: Did you notice any similarity between this list and the contents of Nassim Nicholas Taleb’s Incerto? That’s because the Incerto is, too, about how to spot and get rid of proxies. Fooled By Randomness is about spurious correlations between proxies and real metrics; The Black Swan is about frequencies used as proxies for impact; Antifragile is about averages used as a proxy for distributions; Skin In The Game is about incentives (and proxies are incentives). ]]>