Mental models for leadership & managing people

The mental models

• Incentives

  People do what they're rewarded to do, not what you ask them to do — so a leader's real job is aligning the incentives. If the bonus, promotion path, or recognition rewards the wrong behaviour, no vision statement or pep talk will override it.

• Second-order thinking

  Every management decision ripples: a rushed layoff trims this quarter's costs but reshapes the survivors' trust, precedent, and morale for years. Before you act, ask 'and then what?' about how the team and its culture will respond, not just the immediate result.

• The Peter principle

  People get promoted for excelling at their current job, so your best engineer becomes a struggling manager — stuck at the level they can no longer do well. Managing well means promoting for the skills the new role actually needs, and building senior tracks that don't force everyone into management.

• Goodhart's law

  The moment you turn a measure into a target — tickets closed, calls per hour, lines of code — people optimise the number and the real goal quietly rots. Use metrics to understand your team, but expect any measure to be gamed once it drives rewards.

• Skin in the game

  Leaders who share the downside rather than just delegate the risk earn trust and make sharper calls. If you set the plan, own its consequences alongside the team instead of keeping the credit for wins and passing the blame for losses.

• Regression to the mean

  An exceptional performance is usually followed by a more ordinary one no matter what you do — which is why praise seems to 'jinx' people and a harsh word after a bad month seems to 'work.' Managers who miss this over-credit their criticism and undervalue their encouragement.

• Dunbar's number

  There's a hard cognitive ceiling — roughly 150 — on how many people one person can genuinely know. It's why teams must split, why layers of management emerge, and why a leader simply cannot hold every relationship personally past a certain org size.

• Bottleneck

  A team's output is capped by its single tightest constraint, so a leader's highest-leverage move is finding and relieving that one bottleneck rather than spreading effort evenly. And the most common bottleneck is the manager themselves, when every decision has to route through them.

• Leverage

  A leader's real output isn't what they do personally but what they get done through others — delegation, hiring, and systems are the long levers that move far more than individual effort. The trap is clinging to the individual-contributor work and refusing to let go of the small stuff.

• The Eisenhower matrix

  Sort work by urgent-versus-important: do the urgent-and-important now, schedule the important-but-not-urgent, delegate the urgent-but-unimportant, and drop the rest. Leading people well means pushing that 'delegate' quadrant onto the team instead of drowning in reactive urgency.

• Chesterton's fence

  When you inherit a team, don't rip out a process, ritual, or rule until you understand why it exists — the 'pointless' weekly meeting may be the only thing keeping two functions in sync. Ask what the fence was protecting before you tear it down.

Biases that trip up leadership & managing people

• Fundamental attribution error

  When a report drops the ball you blame their character — 'careless,' 'not committed' — yet blame the situation when you slip. That double standard poisons feedback and fixes the wrong thing, because the missed deadline was often a broken process, not a broken person.

• Self-serving bias

  It's tempting to claim the team's wins as your leadership and pin the losses on the market, the deadline, or a weak hire. Nothing corrodes trust faster than a manager who hoards credit and outsources blame.

• Halo effect

  One vivid trait — polish in a meeting, a single big win, sheer confidence — bleeds into your whole read of a person, so you promote the charismatic over the quietly excellent and end up rating the halo instead of the work.

• Authority bias

  As the boss, your offhand remark carries weight you don't feel, and people defer to your title rather than pressure-test your idea. The result is unchallenged bad calls — you have to actively invite dissent or your seniority silences the room.

• The curse of knowledge

  Once you know how to do the job you can't imagine not knowing it, so your instructions skip steps, your delegation stays vague, and new hires flounder on things that feel 'obvious' to you but were never actually explained.

• Pygmalion effect

  The expectations you hold become self-fulfilling: back someone and they rise to it; quietly write them off and your lowered bar drags their performance down. Managers hand out this prophecy constantly without noticing they're doing it.

• In-group bias

  You extend trust, the benefit of the doubt, and airtime to 'your people' while eyeing new hires and other teams with suspicion — breeding favouritism, silos, and unfair reviews that the in-group never sees as unfair.

Mental models for other work