https://luca-dellanna.com People management and risk management. Sat, 07 Nov 2026 08:01:34 GMT https://validator.w3.org/feed/docs/rss2.html https://github.com/jpmonette/feed en All rights reserved 2026 https://luca-dellanna.com/posts/chief-decision-officer https://luca-dellanna.com/posts/chief-decision-officer Tue, 03 Nov 2026 00:00:00 GMT https://luca-dellanna.com/posts/long-term-iterated-games https://luca-dellanna.com/posts/long-term-iterated-games Thu, 05 Nov 2026 00:00:00 GMT https://luca-dellanna.com/posts/birthday-cake-premortem https://luca-dellanna.com/posts/birthday-cake-premortem Wed, 28 Oct 2026 00:00:00 GMT https://luca-dellanna.com/posts/helping-employees-grow https://luca-dellanna.com/posts/helping-employees-grow Wed, 14 Oct 2026 00:00:00 GMT https://luca-dellanna.com/posts/turin-tables https://luca-dellanna.com/posts/turin-tables Tue, 06 Oct 2026 00:00:00 GMT https://luca-dellanna.com/posts/soul-of-art-keeps-moving https://luca-dellanna.com/posts/soul-of-art-keeps-moving Wed, 30 Sep 2026 00:00:00 GMT https://luca-dellanna.com/posts/human-alignment https://luca-dellanna.com/posts/human-alignment Wed, 23 Sep 2026 00:00:00 GMT https://luca-dellanna.com/posts/incentives-and-distortions https://luca-dellanna.com/posts/incentives-and-distortions Wed, 23 Sep 2026 00:00:00 GMT https://luca-dellanna.com/posts/fix-politics-using-ai https://luca-dellanna.com/posts/fix-politics-using-ai Thu, 17 Sep 2026 00:00:00 GMT https://luca-dellanna.com/posts/engineers-not-activists https://luca-dellanna.com/posts/engineers-not-activists Wed, 16 Sep 2026 00:00:00 GMT https://luca-dellanna.com/posts/training-designed-for-the-trainer https://luca-dellanna.com/posts/training-designed-for-the-trainer Sun, 06 Sep 2026 00:00:00 GMT https://luca-dellanna.com/posts/cheaper-and-better https://luca-dellanna.com/posts/cheaper-and-better Sat, 29 Aug 2026 00:00:00 GMT https://luca-dellanna.com/posts/profit-per-person-per-use https://luca-dellanna.com/posts/profit-per-person-per-use Thu, 03 Sep 2026 00:00:00 GMT https://luca-dellanna.com/posts/pareto-frontier https://luca-dellanna.com/posts/pareto-frontier Wed, 19 Aug 2026 00:00:00 GMT Every time I participate in a high school reunion, I have a choice in how I compare myself to others. One option is to notice that, among my former classmates, there is at least one who earns more than I do, at least one who has more free time than I do, and so on. Another option is to notice that, among my former classmates who earn at least as much as I do, I'm the one with the most free time. And among those who have at least as much free time as I do, I'm the one who earns the most. Learning to view success through this lens is highly beneficial. For example, do you want to be as successful as possible, or do you want to be as successful as possible while always having dinner with your family? Then, you shouldn't compare yourself with anyone who does not also spend all their dinners with their family. **Never compare yourself with someone who has different priorities or is optimizing for different objectives or for a shorter time horizon.** Doing so leads to unhappiness and/or making counterproductive decisions. ## The Pareto Frontier Just now, I presented two ways to compare yourself to others. 1. "There is someone who earns more than I do and someone with more free time than I do." 2. "Among those who earn as much as I do, I have the most free time, and among those with as much free time as I do, I earn the most." The second type of evaluation is called the _Pareto Frontier_, named after the Italian mathematician Vilfredo Pareto (the same one who invented the _Pareto Rule_, also called the "80/20 rule": 20% of your customers are responsible for 80% of your problems, and so on). In particular, I am on the Pareto Frontier if, given two metrics, A and B, among everyone who has at least as much A as I do, I have the most B, and among everyone who has at least as much B as I do, I have the most A.  The Pareto Frontier is also useful for comparing options. For example, without knowing about the Pareto Frontier, you might compare jobs by asking yourself which one is the highest paying. Instead, by using the Pareto Frontier, you might ask yourself better questions, such as, what's the highest-paying job among those that enable you to spend enough time with your family? The Pareto Frontier isn't a revolutionary concept; many of us already make decisions using a similar mental process. However, giving it a name and making it explicit is useful because it shows that choosing a job that's not the highest paying is not necessarily a lazy or suboptimal choice. If it's on the Pareto Frontier of what matters to you, it is an ambitious and optimal choice. ## How the Pareto Frontier helps you win long-term games Single-metric comparisons such as "who earns more" are toxic because they lead to feelings of inferiority and/or races to the bottom. Conversely, comparing yourself to others using multi-metric comparisons such as the Pareto Frontier is a relatively healthy way to assess whether you're being ambitious enough. If you are at the Pareto Frontier in the areas that matter to you, you're probably doing fine (as long as being there isn't the result of short-term optimization). Conversely, if you are far from your Pareto Frontier, you might have been too lazy or conservative. But perhaps more importantly, defining success in terms of Pareto frontiers helps you avoid the temptation to follow those who take excessive risks or make sacrifices you're not willing to take. **Instead of aiming to be the most successful, aim to be the most successful among those who play long-term games and are not willing to take excessive risks or make excessive sacrifices.** Your distribution of outcomes will improve. --- _This is an excerpt from my book [Winning Long-Term Games](https://luca-dellanna.com/books/winning-long-term-games)._ ]]> https://luca-dellanna.com/posts/virtues-based-values-based-cultures https://luca-dellanna.com/posts/virtues-based-values-based-cultures Sun, 16 Aug 2026 00:00:00 GMT https://luca-dellanna.com/posts/culture-outlast https://luca-dellanna.com/posts/culture-outlast Sun, 09 Aug 2026 00:00:00 GMT https://luca-dellanna.com/posts/school-designed-for https://luca-dellanna.com/posts/school-designed-for Sun, 19 Jul 2026 00:00:00 GMT https://luca-dellanna.com/posts/handling-poor-performers-and-missed-deadlines https://luca-dellanna.com/posts/handling-poor-performers-and-missed-deadlines Wed, 08 Jul 2026 00:00:00 GMT https://luca-dellanna.com/posts/leading-ai-adoption https://luca-dellanna.com/posts/leading-ai-adoption Wed, 08 Jul 2026 00:00:00 GMT As a rule of thumb, if your training material is generic enough so that it could be given to a project manager, a salesperson, and an engineer alike, it is not specific and relevant enough to drive adoption. ### Mistake 2: They treated supervisors as bystanders People do not change how they work because the CEO asked them to in a company-wide email. They change when their direct supervisor visibly cares. If the supervisor doesn't use AI, the team won't use it either. If the supervisor doesn't care, daily, about whether their people use it, the team won't care either. This is not because supervisors are cheerleaders for corporate initiatives. It is because they are the most visible person in a team and they control the local standards of work: what counts as good enough, what gets reviewed, what gets repeated, and what quietly dies after the kickoff meeting. Supervisors are the linchpin of any tool or process adoption, and a rollout that doesn't actively enlist them is a rollout that quietly excludes most of the workforce (everyone except self-starters and, maybe, some early users). Enlisting supervisors means two things: securing their genuine buy-in, and [training them to coach their people](https://luca-dellanna.com/posts/coaching), because you cannot assume they already know how to introduce a new way of working without it coming across as meddling. ### Mistake 3: They trained everyone in group sessions A one-hour training for fifty people looks efficient, with a low cost per head. However, it is mostly wasted, for the reason we saw in Mistake 1: adoption is driven by examples that are concrete and relevant to each person's job, and no group session can be relevant to fifty people at once, unless these people have the same job and are exactly at the same level with regards to AI proficiency. Large sessions can be useful for orientation, risk awareness, or shared vocabulary, but they do not generate adoption. One-on-one sessions (or sessions with two or three people who share the same role and proficiency) are far more effective: you sit with the person, take one of their actual tasks, and work through it with AI together. More importantly, you get the time to address their personal doubts and have them practice under your supervision, something that's impossible to do with large groups. One-on-one sessions look expensive, but the time is not wasted, because each session tends to produce something a group session rarely does: an actual adopter. An hour that converts one person beats an hour that washes over fifty. ### Mistake 4: They tried to automate entire jobs The last mistake is aiming AI at a job's core task, with full automation as the goal. But this is exactly where AI is weakest, where the cost of errors is highest, and where people's professional identity puts up the strongest resistance. Thankfully, jobs are not monoliths. They are ensembles of tasks, and tasks are ensembles of steps. For most jobs, the full core task cannot be delegated to AI yet, while the ancillary tasks around it (administrative work, meeting preparation, pre-mortems, reviewing drafts for errors and likely objections, etc.) are full of low-hanging fruit. The same goes for sub-parts of the core task that can be augmented rather than automated. Smaller bites are easier to make work, easier to verify, and easier to build confidence in. For example: do not ask AI to "own customer support," but use it to summarize long tickets, suggest reply drafts, flag missing information, and convert repeated issues into knowledge-base entries. Or do not ask it to "automate lead generation," but instead ask yourself what parts of lead generation can be automated, such as researching background information for a given person, or helping you draft an email not from scratch (that will sound artificial) but fleshing it out from a few bullet points you provide. In each case, AI is not about replacing the professional judgment at the center of the job. It is about removing friction and tediousness. ### Why mandates do not work Notice that none of the four mistakes covered here is about tools or incentives. That is why responding to low adoption with mandates, dashboards, or bonuses doesn't work well. At most, they can force visible usage, but they cannot create skilled usage. [Incentives can buy more of an existing behavior, but rarely create it from scratch](https://luca-dellanna.com/posts/incentives-vs-habits). The latter requires direct interactions between supervisors and their subordinates. ### The supervisor-led playbook Here is what works instead, in three moves and one boundary. 1. **Start simple, and start with supervisors.** You may not need custom tools integrated with your databases to begin; [a simpler path](https://luca-dellanna.com/posts/ai-faster-adoption) is giving people general-purpose AI assistants and training them to use those on their personal effectiveness, which captures a large share of the available gains at a fraction of the investment. Before anything reaches the teams, secure supervisors' buy-in and train them to coach. 2. **One use case per role, with an engineered first win.** Pick a single ancillary task where AI demonstrably helps that role, prepare good prompts, and have the supervisor coach each person through the first use, one-on-one. There is nothing worse than giving someone a powerful tool with no demonstration: they will try it once, it won't go particularly well, and they will lose interest. Frame the whole effort as you-and-them against an external enemy, such as the workload or the competition, never as "I need you to abandon your way of working." 3. **Build judgment, not just procedures.** Once people use AI, the risk shifts to using it badly: trusting hallucinations, missing errors. The fastest fix is running hypotheticals: list five to ten scenarios and dilemmas, ask your team what they would do, and give feedback while explaining your reasoning. A session of thirty to sixty minutes can provide months' worth of experience, including experience in spotting AI mistakes. 4. **The boundary: do not automate before it works with manual prompting.** Premature automation is expensive twice: first because you spend time integrating and polishing workflows that may not matter, and then because you lock people into brittle processes before they have developed the judgment to use them well. In this context, "automation" includes both technical automation and skill automation: agents, templates, custom GPTs, workflow integrations, playbooks, and anything else that turns a human judgment into a repeatable procedure. Do not start with these. Instead, first make the task work manually with prompts, examples, and supervision, and only afterward, codify what works. Otherwise you are not scaling a capability; you are crystallizing an untested guess. ### Measure output, not usage One warning about goals. If the goal you set (or the metric you track) is AI usage, you will get busywork usage: prompts performed for the dashboard, tools opened to be seen using them, and no improvement in results. The goal is increased output: faster turnaround, fewer errors, less rework, more capacity for the work that matters. Usage is a means, and it follows naturally when the use cases are well chosen. This is also what eventually moves resistant users: not a metric telling them they must, but visible results making non-usage the locally strange choice. ### The prize includes upskilling, not just automation Everything above gets AI used. But the biggest return is not (just) the time saved by automating work; it is [using AI to upskill your team](https://luca-dellanna.com/posts/automation-vs-upskilling). AI can codify what your best people do (their standards, checklists, and playbooks) and make it available to everyone else as real-time feedback and quality control: drafts reviewed against your rubric, gaps flagged, improvements suggested, with the human still accountable for what ships. Done well, this is on-the-job training that brings everyone closer to the skill level of your best employee. The floor rises, and outcomes become consistent. The organizations that win with AI won't be the ones that automate the most tasks; they'll be the ones that treat AI as a training-and-quality system. And a team adopted this way compounds: the skills and judgment your people build ride every improvement in the models. ### Want to get better at using AI yourself? This post is about leading a team to adoption. If you want to raise your own skills as an individual user, read [Seven Ways to Get Better Results from LLMs at Work](https://luca-dellanna.com/posts/llm-tips-work) for office tasks, [The Goldfish Student](https://luca-dellanna.com/posts/goldfish-student-ai-context) for giving AI the right context, and [Contracts](https://luca-dellanna.com/posts/skills-and-contracts) for getting consistent, verifiable outputs on recurring tasks. ### Want help putting this into practice? Getting AI genuinely adopted (and used as a training-and-quality system rather than a box-ticking exercise) is one of the most common problems leaders bring to me. If you rolled out AI and usage is low, or usage is high but results aren't moving, that is the kind of problem I work on with leaders. [Here is how to start](https://luca-dellanna.com/advisory). ]]> https://luca-dellanna.com/posts/manager-firefighting-loop https://luca-dellanna.com/posts/manager-firefighting-loop Wed, 08 Jul 2026 00:00:00 GMT https://luca-dellanna.com/posts/perks-dont-retain-people https://luca-dellanna.com/posts/perks-dont-retain-people Wed, 08 Jul 2026 00:00:00 GMT https://luca-dellanna.com/posts/soccer-academies-job-distortions https://luca-dellanna.com/posts/soccer-academies-job-distortions Wed, 08 Jul 2026 00:00:00 GMT https://luca-dellanna.com/posts/hire-outcomes-habits https://luca-dellanna.com/posts/hire-outcomes-habits Wed, 08 Jul 2026 00:00:00 GMT https://luca-dellanna.com/posts/remote-teams-clarity-feedback https://luca-dellanna.com/posts/remote-teams-clarity-feedback Wed, 08 Jul 2026 00:00:00 GMT https://luca-dellanna.com/posts/insurance-warning-signal https://luca-dellanna.com/posts/insurance-warning-signal Sun, 05 Jul 2026 00:00:00 GMT https://luca-dellanna.com/posts/team-coaching https://luca-dellanna.com/posts/team-coaching Sun, 28 Jun 2026 00:00:00 GMT ### Repetition, feedback, and reinforcement Once someone can do a thing competently, three levers turn it into a habit. The first is **repetition.** Keep asking the same question until it becomes the question they ask themselves. A manager who asks "Is there any way we can do this faster or cheaper?" in every review eventually instills that question in the team's heads. The second is **feedback.** If someone asks a question or surfaces a problem and gets ignored or blamed for it, they will learn that the next time they have a question or see a problem, they should pretend everything is fine. The third is **immediate reinforcement.** If you ask someone to learn something, and they learn it, and you don't notice, the lesson they will take is that the next time you ask them to do something, they should put less effort into it. This is not just a question of feeling seen; it is also that not noticing that something got done signals it was not important. ### Case study: coaching better emails Let's put into practice what we have seen so far with the help of a concrete case study: coaching someone to write clearer emails. Instead of announcing a training, wait until an email with clear room for improvement lands in your inbox. In your next one-on-one, mention it and pick one thing to fix, not five. If the email was not concise, do not just say "be more concise," because they probably believe they are already concise, *by their standards*. Instead, show them *your standards*: put the email in front of them and ask them to strike out everything unnecessary, then confirm that what is left reads more clearly. Finally, watch the next few emails, and when they get it right, let them know you noticed. That small loop, observing, teaching one thing specifically, letting them practice, and reinforcing, carries the same principles we just saw: competence before habit, repetition, and immediate feedback. *(You can find my full guide on email coaching, with exercises for conciseness, clarity, and directness, in [How to coach your team to write more effective emails](https://luca-dellanna.com/posts/email-coaching).)* ### Track growth with soft metrics, discussed often Many managers rely on hard metrics to track performance, such as contracts signed or invoices processed, but soft metrics are often more useful than hard ones when it comes to building skills. Take a basketball player who struggles to score. A hard metric like points per game helps only if the limitation is effort, but if the problem is shot selection, that target pushes them to take worse shots to hit the number. A soft metric like "shot quality," discussed after games, addresses the actual skill gap. "But Luca, soft metrics are subjective." That's only a problem if you do not discuss them frequently. Do it only once a quarter, and you may create defensiveness and disagreement. But if you discuss them frequently and informally, say at the end of a one-on-one on another topic, it produces an open, constructive conversation. Plus, the more you discuss them, the more your subordinate's judgment of soft metrics will match yours. There is no need to pretend soft metrics are objective; instead, anchor them in examples: "This is a weak customer update, this is an acceptable one, this is a strong one, and here is why." These metrics also help *you* get feedback on your coaching itself. If your coaching is working, the soft metric should be moving: the customer updates getting sharper, the questions getting better. A soft metric that does not move after weeks of coaching is telling you something, either that the coaching is not landing or that the gap was never skill in the first place. The trajectory of the soft metric is how you know whether any of this is working. *(More about soft metrics in my old essay, [How to use soft metrics to develop your team's skills](https://luca-dellanna.com/posts/soft-metrics).)* ### What about AI? Remember the three requirements that made group training tricky (meet people at their level, work on their real tasks, and give feedback)? They were hard because they did not scale. A human coach can deliver all three, but only to one or two people at a time. AI can help relax this constraint. Beyond automating tasks where it makes sense, AI can act as a training and quality system: it can review a sample of outbound work against your standards, flag what is missing, and suggest improvements, giving people immediate feedback until the standard becomes a habit. It meets each person at their own level, on the specific task in front of them, with feedback delivered the moment the work is done, and it does this for everyone at once. The three requirements that were impossible to satisfy at scale become cheap. For example, if a manager has defined what a clear customer-update email looks like, AI can review drafts against that standard, flag missing context, ask whether the recommendation is explicit, and suggest a tighter version before the email is sent. You can also have your best performer codify what they do into a reusable standard that the tool applies for everyone else, which raises the floor toward the level of your strongest person. This is the same move as having a supervisor coach against an expert's standard, run at the scale of the whole team and at the speed of every draft. **That said, AI should only complement a supervisor's coaching, not replace it.** AI can provide the repetitive, scalable part of coaching, checking work against a known standard and returning immediate feedback. But a human is needed to set the standard in the first place, judge the work AI never sees, coach the things that have no written rubric, and decide what is worth getting good at. *(More on this in my essay [Automation vs upskilling](https://luca-dellanna.com/posts/automation-vs-upskilling).)* ### Conclusion Skills and capabilities aren't built with group training sessions, but by supervisors who coach on real work, build competence before expecting a habit, reinforce the right behavior with repetition and feedback, and keep skills visible through frequent, low-pressure conversations. If you want help turning this into how your organization develops people, diagnosing where skills stall, training your managers to coach, and setting up the metrics that track growth, that is one of the things I do professionally. [Here is how to start](https://luca-dellanna.com/advisory). ]]> https://luca-dellanna.com/posts/wealth-preferences-culture https://luca-dellanna.com/posts/wealth-preferences-culture Thu, 11 Jun 2026 00:00:00 GMT https://luca-dellanna.com/posts/goldfish-student-ai-context https://luca-dellanna.com/posts/goldfish-student-ai-context Mon, 01 Jun 2026 00:00:00 GMT The more information you add, the harder it becomes for the AI to keep all the important information simultaneously present while reasoning. This does not mean that longer prompts are always worse. Sometimes a complex task genuinely requires a lot of context. But it does mean that every sentence should justify the space it occupies in the goldfish's memory. In many cases, a one-page prompt that clearly states the objective, constraints, and relevant background will outperform a ten-page prompt that includes everything remotely related to the task. When in doubt, ask yourself: if the goldfish could only keep ten pieces of information on its desk, which ten would I choose? Then give it those. Or, if you do not want to do the triaging work yourself, ask the AI to do it. Give it a clear purpose and criteria, such as: "Here is a collection of documents about my company. Retain only the information that will be relevant to writing a marketing email for this product and this customer." ## Practical Techniques A useful technique is to separate remembering from reasoning. The reason is that each task uses working memory space. If the AI is trying to remember, summarize, search, and reason simultaneously, those tasks compete with one another. Instead of asking, "Read these ten documents and tell me what strategy I should follow," first ask the AI to extract the relevant information, and only then, ask it to build a strategy from that information. Separating the two steps often produces better results. Another technique is to fully separate the ingestion of resources from the answering. First, give the AI a book and ask it to take notes on specific topics. Then, ask it to show you those notes, and review a sample of them. This step is important because it lets you verify the quality and depth of the notes. If the notes are superficial, the answers will be superficial too. Only after you have validated the notes should you ask questions based on them. A more advanced version combines several techniques: 1. Tell the AI that you are about to give it some reference information, but before you do so, tell it what the information will be used for, so that it knows what to retain. 2. Give the AI one or more books and, for each book, ask it to create both a high-level summary and more detailed notes for each chapter. 3. Ask it to show you a sample chapter note so you can evaluate the quality of the note-taking. 4. Ask your question and instruct it to answer by consulting the summaries first and then digging into the more detailed notes when necessary. The idea is that the AI uses the book summaries as a routing table to know which chapter notes to dig into before answering each of your questions. This ensures that only relevant information occupies its working memory. A further technique is to use multiple agents. This is equivalent to the goldfish recruiting more goldfish and assigning a subtask to each. One reads the documents. Another analyzes them. A third reviews the answer. Each goldfish has fewer things to manage and can focus on a smaller, more relevant subset of information. Note that some of these techniques require AI tools that can create and manage files, such as Claude Code, Claude Projects, or Codex, and are therefore unavailable in a simple web chat interface. If all you have is a web chat, you can still approximate the process by asking the AI to create notes, saving them, and then uploading those notes as the basis for future conversations. For a complementary set of techniques on getting better results from AI, see [5 Techniques for Better AI Outcomes](https://luca-dellanna.com/posts/5-techniques-for-better-ai-outcomes). ## Final Thoughts A common mistake is to assume that AI behaves like a perfect reader that remembers everything it has seen. A more useful mental model is the goldfish student: intelligent, capable, and diligent, but only able to keep a limited amount of information in its working memory at any time. Once you understand that, many best practices become obvious. Give it a purpose before giving it information. Help it take notes. Review those notes. Break large tasks into stages. And whenever possible, reduce the amount of information competing for space in the working memory. ]]> https://luca-dellanna.com/posts/skills-and-contracts https://luca-dellanna.com/posts/skills-and-contracts Sat, 23 May 2026 00:00:00 GMT https://luca-dellanna.com/posts/llm-tips-work https://luca-dellanna.com/posts/llm-tips-work Thu, 21 May 2026 00:00:00 GMT