Level 2 · Module 1: Incentives Run the World · Lesson 5

Designing Incentives That Actually Work

capstonehuman-naturegroups-powercharacter-leadership

Anyone can see when incentives are broken. The harder and more valuable skill is designing incentives that actually work — systems where the behavior you want is also the behavior the system rewards. Good incentive design follows a few principles: reward the outcome you actually want (not a proxy that can be gamed), make the right thing easier than the wrong thing, align all three layers of incentives when possible, and test your design by asking 'How would a clever person game this?' If you can't figure out how to game your system, it might actually be good.

Building On

Incentives shape behavior

The first lesson of this module established the basic principle: people respond to incentives. This capstone asks you to use that principle actively — not just to understand why people behave the way they do, but to design incentive systems that produce the behavior you actually want.

Bad systems make good people do bad things

The previous lesson showed how badly designed incentive systems can corrupt even good people. This lesson flips the question: if bad design produces bad behavior, what does good design look like? How do you build a system that makes the right thing easy and the wrong thing hard?

Three layers of incentives

The three-layer model — material, social, and internal — is your design toolkit. The best incentive systems don't just offer rewards and punishments. They also harness social approval and appeal to people's sense of purpose and pride. A system that aligns all three layers produces the strongest and most lasting behavior change.

Why It Matters

You might think that designing incentive systems is something only adults do — CEOs, principals, coaches. But you design incentive systems all the time. When you're the captain of a team and you decide how to encourage effort, that's incentive design. When you're organizing a group project and figuring out how to make sure everyone does their part, that's incentive design. When you're babysitting and trying to get a five-year-old to clean up their toys, that's incentive design.

The difference between people who are effective at getting things done and people who aren't is often the difference between good and bad incentive design. The effective person understands that telling people to try harder almost never works. What works is making the right behavior the natural, easy, rewarded behavior — and making the wrong behavior costly or difficult.

This is also a capstone lesson, which means it pulls together everything you've learned in this module. You've learned that incentives shape behavior, that incentives can go wrong in predictable ways, that incentives operate on three layers, and that bad systems can corrupt good people. Now you're going to use all of that knowledge to actually design something better.

A Story

Two Coaches, Two Systems

Coach Ramirez and Coach Dixon both ran youth basketball programs in the same league. They had similar players — similar talent, similar ages, similar neighborhoods. But their teams behaved very differently, and the reason was incentive design.

Coach Dixon believed in competition. He posted individual scoring stats after every game. The top three scorers each week got to choose the drill at the next practice. Players who scored the most points over the season got trophies at the end-of-year banquet. His system was clear: score more, get rewarded more.

The results were predictable to anyone who understood incentives. Dixon's players stopped passing the ball. Why pass to a teammate when only your points get tracked? The best scorers took bad shots instead of passing to open teammates, because a bad shot that goes in counts the same on the stat sheet as a good one. Players who were good at defense, rebounding, or passing — things that didn't show up in scoring stats — felt invisible. Some of them quit.

Dixon's team won some games because they had talented scorers. But they also lost games they should have won because nobody played together. And the players who stayed weren't developing into complete basketball players — they were developing into selfish ones.

Coach Ramirez took a different approach. She tracked assists (passes that led to baskets) as prominently as points. She gave weekly recognition for 'plays that helped the team' — which could be a great pass, a crucial rebound, a smart defensive switch, or encouraging a struggling teammate. She gave playing time based on effort and teamwork, not just talent. And she asked the players themselves to vote on who had been the best teammate each week.

Ramirez's system aligned all three layers. The material incentive (playing time, recognition) rewarded team play. The social incentive (teammate votes, public recognition of assists) made unselfish play something the group valued. And the internal incentive was reinforced by the constant message: the purpose of this team is to make each other better, not to make yourself look good.

The results were dramatic. Ramirez's players passed the ball. They played defense. They encouraged each other. Players who weren't natural scorers found ways to contribute and felt valued for it. The team improved faster because everyone was developing all their skills, not just shooting.

By the end of the season, Ramirez's team had a better record than Dixon's — not because they had better individual players, but because their system brought out the best in the players they had. Dixon's system had taken talented individuals and made them worse as a team. Ramirez's system had taken similar individuals and made them better.

The difference wasn't that Ramirez's players were nicer people or that Dixon's were selfish. Put Ramirez's players in Dixon's system, and they'd start ball-hogging too. Put Dixon's players in Ramirez's system, and they'd start passing. The difference was the design of the incentive system — and the designer was the coach.

Vocabulary

Incentive design
The deliberate creation of rules, rewards, and structures that encourage the behavior you actually want. Good incentive design makes the right thing easy and rewarding. Bad incentive design accidentally encourages the wrong behavior.
Proxy measure
A substitute measurement used when the thing you really care about is hard to measure directly. Points scored are a proxy for 'basketball skill' — but they're an imperfect proxy because they miss passing, defense, and teamwork. When people optimize for the proxy instead of the real thing, the system breaks down.
Gaming the system
Finding ways to get the reward without actually doing what the system was designed to encourage. If a system rewards page count, someone will write with bigger margins. If a system rewards test scores, someone will teach only what's on the test. Good incentive design anticipates gaming and makes it difficult.
Alignment
When the behavior the system rewards is the same as the behavior you actually want. A well-aligned system makes doing the right thing and getting rewarded feel like the same action. A misaligned system forces people to choose between doing the right thing and getting rewarded.

Guided Teaching

Start with the two coaches. Ask: 'What exactly did Coach Dixon measure and reward? What behavior did that produce?' Then ask the same for Coach Ramirez. The key insight is that both coaches got exactly what they incentivized — Dixon got individual scoring because he rewarded individual scoring, and Ramirez got teamwork because she rewarded teamwork. Neither coach got surprised by their results. They each designed a system and got the behavior that system produced.

Introduce the concept of proxy measures. Ask: 'Points scored seem like a good way to measure basketball ability. Why didn't it work for Coach Dixon?' Because points are a proxy — they measure one part of basketball but miss defense, passing, rebounding, and leadership. Ask: 'Can you think of other proxy measures that miss important things?' Grades as a proxy for learning, likes as a proxy for quality, money as a proxy for success.

Teach the gaming test. Ask: 'If you were a clever player on Coach Dixon's team, how would you game his system to maximize your rewards?' Take lots of shots, don't pass, play flashy rather than smart. Now ask: 'If you were on Coach Ramirez's team, how would you game her system?' It's much harder to game — because assists require helping teammates, and the teammate vote is hard to fake. This is the design test: a good system is hard to game.

Walk through the design principles. List them: (1) Reward what you actually want, not a proxy. (2) Make the right thing easier than the wrong thing. (3) Align all three layers — material, social, and internal. (4) Ask 'How would someone game this?' Ask the student to evaluate each principle using the two coaches as examples.

Close with the capstone challenge. Present the student with the practice exercise: they're going to redesign a broken incentive system. Ask: 'Now that you've seen how incentive design works, can you take a system that's producing bad behavior and redesign it to produce good behavior?' This is where they apply everything from the module.

Pattern to Notice

Whenever you see a system producing behavior that nobody seems to want, look at what the system actually rewards. The behavior the system produces is almost always the behavior the system incentivizes — even if that's not the behavior anyone intended. The pattern is: (1) someone designs a system with good intentions, (2) the system rewards a proxy or the wrong behavior, (3) people optimize for what's rewarded, not what's intended, (4) everyone complains about the behavior without changing the system. Breaking this pattern requires stepping back and asking: 'What is this system actually rewarding?'

A Good Response

A student who has mastered this capstone can take a broken incentive system, diagnose why it's producing bad behavior, and redesign it to produce better behavior. They can explain the four design principles, identify proxy measures and their limitations, and test their design by asking how it could be gamed. They understand that the designer of an incentive system bears responsibility for the behavior that system produces.

Moral Thread

Responsibility

Designing good incentives is an act of responsibility — because when you set up the rules that shape other people's behavior, you are partly responsible for the behavior those rules produce. The careless designer who creates a system full of perverse incentives is not innocent when people behave badly in that system. Understanding how incentives work gives you the power to build systems that bring out the best in people instead of the worst. That power comes with responsibility.

Misuse Warning

This lesson could be misused to believe that all behavior is just a response to incentives — that people are simply machines who do whatever the system rewards. That's too simple. People have free will, conscience, and character. Good people sometimes resist bad systems, and bad people sometimes corrupt good systems. Incentive design is powerful, but it doesn't control everything. The lesson is that systems matter enormously — not that they're all that matters.

For Discussion

  1. 1.If you put Coach Ramirez's best players on Coach Dixon's team, what would happen to their behavior? What does that tell you about the power of system design?
  2. 2.What is a proxy measure, and why do proxy measures often produce gaming instead of genuine improvement?
  3. 3.Can you think of an incentive system in your life that works well — where the behavior it rewards is the behavior you actually want?
  4. 4.What are the four principles of good incentive design? Which one do you think is most important?
  5. 5.If you were designing the rules for a group project at school, how would you make sure everyone does their fair share? What would your system reward?

Practice

Redesign a Broken System

  1. 1.Choose a real incentive system that you think is broken — something at school, on a team, in your community, or even a game you play. It should be a system where the behavior people actually produce is not the behavior someone intended.
  2. 2.Diagnose the problem. Answer: (1) What behavior does the system currently reward? (2) What behavior did the designer probably want? (3) What's the gap between the two? (4) Is there a proxy measure being gamed?
  3. 3.Redesign the system. Create new rules, rewards, or structures that would produce better behavior. For each change, explain which of the four design principles you're applying.
  4. 4.Test your design. Ask: 'How would a clever person game my new system?' If you can find a way to game it, try to close the loophole.
  5. 5.Predict the results. Write 3-4 sentences describing what you think would happen if your redesigned system were actually put in place. Be honest about what might go wrong as well as what might go right.
  6. 6.Present your redesign to a parent or teacher and ask them to try to find weaknesses in it. A good system should survive tough questions.

Memory Questions

  1. 1.What are the four principles of good incentive design?
  2. 2.Why did Coach Dixon's system produce selfish play and Coach Ramirez's system produce teamwork?
  3. 3.What is a proxy measure, and what happens when people optimize for the proxy instead of the real thing?
  4. 4.What does it mean to 'game a system,' and why is the gaming test important for incentive design?
  5. 5.Why does the person who designs an incentive system bear some responsibility for the behavior it produces?

A Note for Parents

This capstone lesson asks your child to move from understanding incentives to designing them — a significant intellectual leap. The practice exercise is the heart of the lesson: your child will choose a broken system, diagnose its incentive failures, and redesign it. Your role is to be a thoughtful critic — ask 'How could someone game your new system?' and help them see weaknesses they missed. Don't be too easy on them. The ability to design incentive systems that survive stress-testing is a skill they'll use for the rest of their lives, whether they're running a team, managing a project, raising children, or leading an organization. The harder you push them now, the better their designs will get.

Share This Lesson

Found this useful? Pass it along to another family walking the same road.