How to get incentives wrong

What do you really want? is a weekly conversation to help you figure out what you really want and how to get it at work and in life.


Happy Mothers day! You know who’s good at incentives? Moms. My mom was always great at motivating me to work hard and do my best (even though I often pushed back).

But organizations (especially governments) have a tough time with incentives. You get what you reward, yet it's difficult to anticipate all the ways rewards alter behavior.

This week my son saw that I have a goal of writing 1000 words a day. He smiled, "Oh! You could just type one letter + space + one letter + space over and over and do that really fast!"

I raised my eyebrows and waited for him to think that through a bit. Eventually, he admitted, "But ... that would be cheating." Even a 10-year-old understands that intent matters.

I'm not writing 1000 words a day only to check that off my list and move on. My goal is to share meaningful ideas to help others. 1000 words just happens to be one way I'm measuring daily progress.


3 real-world examples of incentives gone wrong:

  1. Paycheck Protection Program

    Banks were enlisted to help with PPP loans intended to help small businesses. But they were compensated based on the amount of money given. If you could give up to $10m loans and received a % back, which size of loan would you prioritize?

    Not rocket science. Banks sorted the list by deal size and gave the money away to all of the biggest businesses. The first allocation of funds quickly ran out. Businesses like Boeing, Shake Shack, and Zagg were awarded loans (many large businesses have now returned them). This made for a PR mess and delayed help to the businesses that are struggling.

  2. Anger about Employment

    A spa owner had been going without a paycheck herself and was excited to let her employees know that she had received financing to still pay them. Their response? Anger.

    The anger came from employees who’d determined they’d make more money by collecting unemployment benefits than their normal paychecks.

    “It’s a windfall they see coming,” Black-Lewis said of unemployment. “In their mind, I took it away.”

    “I couldn’t believe it,” she added. “On what planet am I competing with unemployment?”

    What was supposed to help, made a mess for an employer trying to do the right thing. And what should be "getting people back on their feet" was actually incentivizing people to not work.

  3. Broken in pieces

    One of my favorite examples of this is the story of an archaeological discovery in the 1940's:

    ... a team led by Ralph von Koenigswald had found another group of early humans which became known as the Solo People, from the site of their discovery on the Solo River at Ngandong. Koenigswald’s discoveries might have been more impressive still but for a tactical error that was realized too late. He had offered locals 10 cents for every piece of hominid bone they could come up with, then discovered to his horror that they had been enthusiastically smashing large pieces into small ones to maximize their income.


How do you avoid greedy businesses, people angry about employment, and broken bones? Two specific actions:

  1. Get people bought into the why

    Understanding the why changes behavior more than trying to dictate every possibility. Before you explain what needs to happen, talk about why.

  2. Prototype

    We are horrible at predicting the future. All 3 examples had poor consequences. By the time they were discovered, some damage was already done. Prototyping is a powerful way to learn early in the process.

    For non-product issues (like all 3 scenarios above) a role-play is especially helpful. This could have been a quick mock run of the intended system with people assuming the roles. Or even the first real example being observed before scaling. Both would have helped the teams responsible for deciding to learn sooner what assumptions they had wrong.

Purpose matters. Prototyping brings learning.

How have you seen real understanding and learning something early impact behavior?


This week I published about how product managers need to be aware of cost before the commit to build and how to create space for focused work: