A sales contest for the month. Whoever sells the most of the new item gets a gift card. Employee of the month on a plaque by the host stand. A points platform where cast members earn badges. A bonus tied to hitting a labor number.
Every one of those arrived from outside your operation, and every one of them is now telling your cast what this operation actually rewards. Which means every one of them is overruling whatever you said your standard was.
What An Incentive Program Actually Is #
An incentive is a statement about what this operation values, made in the only language that is never disbelieved: money and standing.
Your cast does not read your wall. They read what gets paid and who gets praised. That is the real reward structure, and a purchased program installs one without anybody deciding what it should be.
The inference: rewarded behavior increases; therefore rewarding good behavior produces a better operation.
The first half is true and it is the problem. Rewarded behavior increases whether or not it was the behavior you wanted, and a program built to work in any restaurant necessarily rewards the things that are countable everywhere.
It Rewards What Is Countable, Not What Carries Load #
Countable in any operation: units sold, check average, labor percentage, upsells attempted, covers turned, review count.
Not countable in any operation, and therefore not in any program you can buy: the cast member who took the difficult table so somebody else did not have to. The one who caught a problem before the Guest saw it. The one who told the truth about a mistake that would have gone unnoticed. The one who made a new hire competent in half the usual time. The one who held pace when the kitchen was a station short. The one who remembered a Guest’s circumstance and handled it without being asked.
That second list is your actual operation. None of it appears in a program, so a purchased incentive systematically pays for the first list and ignores the second, and your cast draws the obvious conclusion about what counts here.
A Contest Produces One Winner And A Building Full Of Losers #
This is the part nobody says out loud, and it is the most damaging thing in the category.
A contest is a mechanism for ranking people against each other. It has one winner by design. Everybody else is told, publicly and on a schedule, that they were not it.
Count them. Twenty cast members, one gift card, nineteen people who came in and did the work and got sorted below somebody else. Run it monthly and you have built a machine that manufactures that result twelve times a year, and you are paying for it.
The ones who never win stop trying by the third cycle, because they have correctly read that the outcome was decided by the section they were assigned, the daypart they work, or the station they are on. Now the program has taught them that effort does not change their standing here.
The one who wins gets something too, and it is not all good. The cast knows who the operation just ranked above them, and nothing about that helps the person who has to work next to them on Friday.
And the whole thing is optional. Nothing about paying people well, recognizing real work, or building capability requires ranking your cast against each other. The ranking was imported with the program, it is the part that does the damage, and it does not have to be there at all.
The work in a dining room is shared. Everybody’s shift depends on everybody else’s. A contest takes shared production and scores it individually, which means it is lying about how the operation actually works, and then paying somebody for the lie.
Six Ways It Damages The Operation #
It converts a cast into competitors. A contest between individuals means the correct move is to stop helping the person next to you, and helping each other is the whole basis of a shift. You paid money to install that.
It teaches selling over reading. Upsell contests put a cast member’s income against the Guest’s actual want. Guests can feel it. What you have installed is a person working an angle at a table where somebody came to eat.
It buys behavior that stops when the payment stops. Behavior driven by a contest reverts the month the contest ends, and now your cast has learned that the operation’s interest in that behavior was temporary.
It punishes the section, the station, and the shift. The Tuesday lunch server cannot win a volume contest. The person on the hardest station cannot win a speed one. Over time the same people win, everybody knows why, and the program is now a monthly reminder of who the schedule favors.
It overrules your ruling at the exact moment it matters. You ruled that a cast member should take the time a table needs. The contest pays for turns. At seven-forty on a Friday your cast runs the contest, because the contest pays.
Recognition with no criteria reads as favoritism. Employee of the month with no stated standard tells your cast that standing here is awarded, not earned. That is worse than no program.
The Legitimate Thing Underneath #
The need is real. People should get more when they produce more, capability should be visible, and good work should be named out loud. An operation that never does any of that runs on goodwill until the goodwill runs out.
The answer is a reward structure you designed against your own standard, not a program you bought. Same need, opposite instrument.
What Replaces It #
Build a performance dashboard for the cast. This is the first move and it replaces most of what a contest was pretending to do. Sales, shifts, covers, average check, the standard’s leading indicators, by person and by shift. Posted, current, and visible to everybody without anybody having to ask for it.
It is for the cast, not for you. They see where they stand, including where they stand against the people they work with, and that is the point. A capable person wants to know. What a dashboard gives them is the one thing a monthly contest never did: a position they can move, starting on their next shift.
Here is why it does not produce what a contest produces. Nothing is awarded, so nobody is declared anything. Nothing closes, so nobody is out. There is no single winner, which means there is no fixed set of everybody else. A number that is fourth this week can be first next week, and every person looking at that board knows it. That is information doing its own work, and it is a completely different mechanism than a prize handed out once a month.
Make the numbers comparable or the board lies. A Tuesday lunch server reading himself against a Friday dinner server is not seeing capability, he is seeing the schedule, and that is the exact defect that makes a contest rot. Post the daypart, the section, and the cover count alongside the number so everybody is being read against the job they actually worked. Per-Guest and per-shift figures do this better than totals.
A board is not a development instrument, and it does not replace one. Nobody gets better because a number went up where they could see it. Development is coaching, reps, and standard-setting done in person, and it still has to happen. The board is the read that tells everybody where they are standing while that work goes on.
What it produces is a cast that knows where the operation stands and where they stand in it. They see a slow Tuesday coming and adjust. They see the shift they worked show up in a number. They stop guessing at whether the operation is doing well and start reading it, which is the beginning of a cast that holds the standard without you in the building.
It also ends your information monopoly, and that is the part most operators are nervous about. Once your cast can see the numbers they can ask real questions, and some will be about decisions you made. That is the cost, it is worth paying, and the operations that pay it get people who think about the business instead of people who think about their section.
Award the better cast members, and award them in the things they actually want. The good shifts. More PTO. The sections they want. First call on the schedule. The right to train, and the right to decide more without asking. These outrank a gift card every time, because they change a person’s week rather than their Friday.
The reason this works where a contest does not is that nothing closes. It is not a monthly title with one holder. It is a standing arrangement: the people producing at the top of the board get treated like it, continuously, and anybody can move into that. The board shows them the distance and the reward shows them it is real.
Say the basis out loud. The cast should be able to point at the board and name why somebody has the Friday. Awarded without a stated basis, the same decision reads as favoritism, and that read costs you more than the reward buys.
Then watch the loop, because this is where it can go wrong. Better shifts produce better numbers, and better numbers earn the better shifts. Read on per-Guest and per-shift figures, that loop compounds in the right direction — the strongest people are on the hardest nights and everybody can see the path. Read on raw totals, it becomes self-confirming: whoever got the Friday first stays on top of the board because it is Friday. Same comparability rule as the board itself, with more riding on it.
Rule what this operation rewards, in writing, before any money moves. Name the behaviors that carry load in your building. That list is the design and it is yours to make.
Pay for capability, not for a month’s activity. A cast member who can run three stations, train a new hire, or hold the standard as the senior person on a shift is worth more permanently. Put it in the pay, not in a contest. Permanent capability, permanent money.
Reward the section and the shift, not the individual, and stop ranking people against each other. The work is shared, so reward it that way. Nobody has to lose for somebody else to be paid well. A dining room is a shared production. Pay it that way and your cast will cover each other without being asked.
Make authority the top of the ladder. The most valuable thing you can give a capable person is the right to decide more without asking. It costs nothing, it is what keeps strong people from leaving, and it is the reward that actually builds the operation while it is being given.
State the criteria for any recognition, out loud, in advance. If you are going to name somebody, say exactly what they did and why it matters here. Named behavior teaches. A plaque does not.
Read the reward structure against your standard, and rebuild what contradicts it. Every target, bonus, contest, and criterion in the building. Anything paying for something your standard does not want comes out. This is the step that makes the rest hold.
Then check reciprocity in both directions. What does a cast member get here that they cannot get down the street, and what are they putting back. A reward structure is only working when both halves are true.
What Changes Tomorrow #
Write down every way somebody in your operation can earn extra money, standing, or preference. Contests, bonuses, recognition, better sections, the good shifts, who gets asked to train, whose opinion gets weight in a pre-shift. All of it, including the informal ones, because those are the loudest.
Now write down the five behaviors you most need on a Friday night. Set the two lists side by side.
Every behavior on the second list with nothing on the first list pointing at it is a behavior your operation says it wants and does not pay for. Every item on the first list pointing at something not on the second list is money you are spending to get something you did not ask for.
Fix one of each this week. Kill the loudest thing paying for the wrong behavior, and put real money or real authority behind the most valuable thing on your list that currently earns nothing.
Your cast will notice both moves within a shift, and they will tell you more about what your operation actually rewards than any program you could buy.
Digging Deeper #
Every term in my framework lives in the Knowledge Base: kb.jeffreysummers.com
Terms used here: [People Constraint], [Cast Contract], [Authority To Execute], [Designed Architecture], [Default Architecture], [Default Gravity], [Measurement Lock-In], [Reciprocity Test], [Sameness Machine], [Ordinary By Design], [Positive Comp], [Cover Blindness], [Assumed Cause]
More of my work on this, all of it in the Knowledge Base:
- 5.TA.1.2 Inside The P&L, The Four Cuts — where the money for this actually comes from
- 4.SH.4a Grow The Floor. Push The Ceiling. — the read a board is for
- 3400 The Operator’s Reps — what actually builds capability
- 3580 Why They Leave, What It Costs, And What The Research Confirms — the real retention math
- 3595 Scheduling Is A Leadership Decision — why better shifts are a real reward
- 3200 The Cast You Can’t Develop From A Dashboard — the board is a read, not development
- Eight Of Your Ten Cast Members Are Not Giving You Their Best — the size of the problem
- Tipping Is Agency — the adjacent argument on Guest-side incentive
