The Peter Principle: 7 Restaurant Promotions That Backfire

The Peter Principle is the observation that people in an organization get promoted based on their current job, until they reach a level where they no longer perform well — and on a restaurant floor that happens faster than almost anywhere else, because the leap from best craftsperson to manager is bigger there than in nearly any other industry.

The pattern is recognizable across the entire industry. Your strongest commis becomes chef de partie. Your fastest server becomes shift leader. Your most reliable bartender gets the keys and the schedule. Six months later the kitchen runs slower than it used to, two people on the team have quit, and the person you promoted is unhappier than when they were simply cooking their own dishes or serving their own tables.

This isn't an exception, and it isn't a case of 'picking the wrong person'. It's a predictable mechanism with a name: the Peter Principle, first described by Laurence J. Peter and Raymond Hull in their 1969 book. The core claim is uncomfortably simple: in every hierarchy, people rise through promotion to the level of their own incompetence — because the only information a promotion is based on is performance in the current role, and that role rarely says anything about the next one.

What long looked like a bitter joke has since been backed by hard evidence. Economists Alan Benson, Danielle Li and Kelly Shue published a large-scale study in the Quarterly Journal of Economics in 2019, using sales data from 214 companies: the employees who sold the most had the highest odds of being promoted to sales manager — and those same top sellers turned out, on average, to be the weakest managers, measured by their team's performance afterward. Companies based promotion decisions almost entirely on sales figures, while the leadership signals that actually predicted success as a manager barely factored in.

This guide walks through seven concrete ways that pattern plays out in a restaurant or a kitchen, with the mechanics behind each one. At the bottom is a test that measures two things separately — technical strength in the current job, and the leadership signals that sit apart from it — so you can see, before you promote someone, which of four zones that person is in today. Everything runs in your own browser: nothing is sent or stored.

Why the leap is bigger in hospitality than anywhere else

In many office jobs, the skills of an execution role and a leadership role partly overlap: a good analyst already understands something about setting priorities and communicating, and that helps when he becomes team lead. In a kitchen or on a restaurant floor, that overlap is much smaller. The skills that make someone the best commis — speed, precision, a memory for a hundred tickets at once, physical endurance through a three-hour service with no letup — are almost the opposite of what a chef de partie needs in order to lead: patience with a slower-learning colleague, staying calm when someone else makes a mistake, building a schedule, having a difficult conversation about a no-show.

The same holds on the floor. The fastest server is often fast because he doesn't let himself get distracted by what the rest of the team is doing — exactly the trait a shift leader must not have, because a shift leader has to constantly notice who needs help, who's falling behind, and who has a difficult guest at their table. These are two separate axes, not a ladder where one automatically leads to the other — and that is exactly the confusion the Peter Principle feeds on: an organization sees one ladder, while in reality there are two independent scales.

The seven mechanisms below aren't a reproach aimed at the owner who chose wrong. They're the predictable outcome of a promotion system that measures only one signal — current performance — while it's actually deciding about a completely different skill. Recognize the pattern before the next promotion, and you can correct it before it costs you a good craftsperson and a functioning team.

The Ultimate Guide Restaurant Staff Guide: 6 Steps to Hire, Train & Retain Hire, train, schedule and keep a brigade in a tight labour market: the complete staffing system for restaurants, from first interview to a culture people stay for. Open the guide

The 7 promotions that go wrong most often

They're listed in the order the pattern usually unfolds: first why the choice itself is already skewed, then why it's so hard to reverse, and finally what you can concretely do before you announce the next promotion.

1. Your best cook becomes head chef — and nobody ever checked whether he can lead

It happens almost automatically: a chef de partie position opens up, and the owner's eye lands on the commis who works fastest, makes the fewest mistakes, and holds up longest through a double shift. That choice feels logical — it is, after all, the best cook in the building — but it tests exactly nothing of what the new role actually requires. Being a chef de partie means splitting tickets across the section, calmly explaining something twice to a younger commis instead of just doing it faster yourself, and staying calm when a different station falls behind, not your own.

This is the core of what Benson, Li and Shue found in their research into sales promotions: the best individual performer becomes, almost automatically, the leading candidate for promotion, while that performance figure predicts little to nothing about how well that same person will lead a team. In their sample of 214 companies, the promoted person's team revenue systematically fell, measured against comparable teams under a different manager — exactly the pattern an owner feels on the floor as 'the kitchen's been slower since he became chef', without being able to point to why.

What makes someone the best craftsperson is largely something different from what a manager needs — an illustrative rendering of the kind of mismatch the sales-promotion research uncovered, not the exact figures from one study.

What stands out in the current job

• Speed under pressure
• Flawless execution
• Endurance through a long service

What a manager needs

• Patience with a slower-learning colleague
• Calm at someone else's mistake
• Plans and follows up without nagging

Neither profile is 'better' — they're two separate skill axes. The problem only arises when an organization treats them as one ladder and measures only the first.

2. Technical skill and leadership are two separate axes, not a ladder

The underlying mistake is a faulty mental model: the assumption that leadership is simply a higher level of the same talent that already makes someone good at their current job — that a chef de partie is 'just a better commis'. That isn't true. Someone who cooks well is good at precision, speed, and repeating a fixed routine until it's second nature. Someone who leads well is good at something structurally different: dividing attention across several people at once, staying patient with a mistake that isn't theirs, and making decisions on incomplete information in the middle of a busy service.

Put those two skills side by side instead of one above the other, and the question you ask yourself changes. Not 'who's best at their own job here', but 'who shows, regardless of how well he handles his own board or his own tables, signs of the second skill' — patience with a slower-learning colleague, calm at someone else's mistake, and the kind of unprompted initiative that keeps a schedule or a briefing on track. The graphic below makes that difference concrete: two entirely separate profiles of the same person.

3. You pay the price on the floor, long before it shows up on paper

A failed promotion costs you in three places at once, and none of them land on an invoice. First, you lose your best craftsperson from the old station — the very commis who worked fastest and most flawlessly is now somewhere else and has to be replaced exactly where he used to excel. Second, engagement drops on the team that inherits the new manager: Gallup's research across hundreds of thousands of teams consistently points to the direct manager as the single biggest factor in how engaged a team is — bigger than pay, bigger than the organization as a whole. Third, the promoted person themselves is at risk: someone struggling in a role they weren't ready for isn't just less effective — they're also considerably more likely to leave on their own, and the same body of research shows the quality of the direct manager is one of the strongest predictors of voluntary turnover, in both directions: of the team that leaves, and of the manager who does.

That's why a failed promotion rarely feels like one obvious event. It feels like a string of small things that happen to coincide — a slightly slower kitchen, two resignations back to back, a chef who's himself thinking about quitting — when in reality it's all one and the same cause. The calculator further down puts a concrete figure on that, based on your own wage cost and your own estimate, rather than a vague gut feeling.

Not on one invoice, but spread across three places at once — which is exactly what makes it so hard for an owner to catch in time.

Best craftsperson gone from the station he excelled at (34%) Team engagement under the new leadership (40%) Risk that the promoted person leaves on their own (26%)

An illustrative split based on the three mechanisms in this article, not a measured percentage from any one specific business. The calculator further down lets you work with your own wage cost and your own estimate.

4. One promotion is hard to walk back — which keeps the wrong choice in place longer than it needs to be

Once someone carries the title of chef, shift leader or team lead, reversing it becomes almost unthinkable for an owner. Not because it's impossible, but because it feels like a double failure: first the wrong choice, then having to openly admit it. Organizational psychologist Barry Staw described this pattern in 1976 as escalation of commitment: the more you've already invested in a decision — time, credibility, the trust you gave the rest of the team — the harder it becomes to reverse it, even when every new signal says you should.

The result is that a promotion that's clearly not working after three months often drags on for a year or more — with a team that in the meantime does feel it isn't working, and a promoted person who also senses they're struggling but sees no dignified way back. The sooner you recognize it, the smaller the step back feels: adjusting a role after six weeks with extra coaching is a very different thing from unwinding it completely after a year. The test further down exists precisely to make that call before the promotion, not after.

5. Training is the difference between a promotion and a gamble

Not every promotion has to go wrong, even when the leadership signals are still weak at the start — provided there's a concrete training plan behind it, rather than the assumption that 'he'll pick it up as he goes'. Benson, Li and Shue found, in the same research, that companies which explicitly weighed leadership indicators alongside performance figures when promoting ended up with structurally better-performing managers and less turnover on their teams — the difference wasn't in who they promoted, it was in what they measured before they did it.

For a restaurant, that means concretely: promoting someone with an explicit training schedule attached — who teaches what, who demonstrates it, when the new manager first works solo — instead of handing over the title and hoping growth follows from experience alone. The site has two tools built exactly for this problem: the training plan splits a new role into what can't wait until the first shift and what follows in phases, and the skills matrix makes visible who on the team can already handle which tasks — including the tasks a manager needs, independent of who already carries the title.

6. Sometimes the best leader isn't your best cook — and overlooking that person is just as costly

The Peter Principle cuts both ways, and the second direction rarely gets any attention: if you promote based on who stands out most through speed or volume, you systematically miss the colleague who works more quietly but does show the other signals — patience with a slower-learning newcomer, initiative on the schedule, the kind of trust where colleagues come to them on their own with a problem. That person rarely tops an 'employee of the month' list, precisely because leadership behavior doesn't always show up in the same numbers as output.

That same research into sales promotions found that companies which also selected explicitly on leadership indicators — alongside performance — ended up promoting people who weren't always the top salesperson on paper, and that this choice worked out better at the team level. For a kitchen or a floor, that means the test further down isn't only there to confirm a promotion you already had in mind — it's also there to surface a name you hadn't considered yourself.

7. What you can concretely check before you announce the title

The six previous points all point to the same practical correction: measure leadership signals separately from performance, before you make a decision that's hard to reverse afterward. That doesn't have to be a formal assessment — these are questions you as an owner or chef can already answer based on what you've watched happen on the floor or in the kitchen over the past months: has this person ever, unprompted, calmly explained something to a colleague, even when it slowed them down? Did they stay calm when someone else made a mistake during a busy service? Do they handle the unglamorous side of the job — a schedule, an order, a briefing — without you having to chase it? And would the team pick this person themselves, if you asked anonymously?

Each of those four questions measures something separate from how fast or flawlessly someone handles their own board or their own tables — and together they form exactly the second axis the Peter Principle overlooks in the first place. The test below turns those four answers into a concrete picture: where this person stands today, in which of four zones, and roughly what a wrong call could cost you if you ignore it.

Test: Is this promotion ready to happen?

Answer the four questions below for the person you're considering promoting, and indicate how strong that person already is in their current job. The test plots those two scores on a quadrant — technical strength against leadership signals — with the real 'Peter Principle zone' explicitly marked: strong in the current job, but without, at least not yet, the signals the next job needs.

Then fill in the wage cost of the team this person would lead, along with your own estimate of how much productivity a wrong match costs and how long it would take before you'd correct it. That last part is deliberately an estimate you set yourself — there's no universal percentage for 'the cost of a failed promotion', and any figure that claims to know one is simply making it up.

Two things worth keeping in mind about your own result. The quadrant is a snapshot, not a verdict — someone in the Peter Principle zone can move to 'ready to promote' within a few months with the right training plan, and that's exactly the point of the test: seeing the problem before the promotion, not after.

And if the result points to 'hidden leader': that's not about passing over the person who's already technically strong, it's about adding a second name to the list you'd otherwise never have considered. Both routes — upskilling someone technically, or giving someone leadership responsibility before they're the strongest technically — cost time. A wrong promotion without that groundwork costs more, as the calculator above shows.

What to do before you announce the next promotion

Nobody fixes seven mechanisms in one week. This order works, because each step makes the next one testable.

Before you have someone in mind — measure the two axes separately

• Run the four questions from the test above for your two or three strongest candidates, not just for whoever comes to mind first.
• Ask yourself who on the team already goes to a particular colleague on their own with a problem — that signal rarely shows up on a performance report.
• Use the skills matrix to see who can already handle which tasks, independent of who carries the title.

Before you announce the title — build the training plan first

• Use the training plan to lay out what can't wait during the first shifts and what follows in phases, instead of handing over the title and hoping growth follows from experience.
• Agree on a concrete trial period — six to eight weeks — with a moment where you explicitly evaluate together, rather than silently waiting it out.
• Work out the cost of a wrong match with the calculator above, so you know what's at stake if you skip the training plan.

If it's not working after the trial period — reverse it early

• Recognize escalation of commitment: the longer you wait to correct course, the bigger the step back feels for everyone — including yourself.
• Talk honestly with the person about whether the role fits, with the craft as an honourable alternative — going back to the station where they excelled isn't a demotion, it's a correction.
• Repeat the test from this article for the next candidate before you decide again, instead of automatically reaching for whoever stands out most right now.

The Peter Principle isn't a judgment on who you chose — it's a blind spot in how promotions usually work

Almost every owner recognizes at least one of the seven patterns above straight away, and that's no coincidence: the Peter Principle isn't a rare exception, it's the standard outcome of a system that measures only one thing — performance in the current job — to make a decision about a completely different skill. That was true in 1969 when Peter and Hull first described it, and it still held up in the large-scale data Benson, Li and Shue analyzed in 2019.

The fix isn't 'promote less' or 'distrust your best people'. It's: measure the two axes separately, before you make a decision that's hard to reverse afterward. Whoever scores strong on both, promote with confidence. Whoever is strong in the current job but still lacks the leadership signals, give a training plan first, not just a title. And whoever already shows the signals without being your best craftsperson deserves a spot on the list you'd otherwise never have considered.

Do that before the next vacancy exists, not after. The test above is meant to make that conversation concrete — with your own team, not with a rule of thumb for 'the industry'.