Alex Weinstein
On Growth, Tech, and Leadership
Gaming the Incentives

game-theory-paper-matrix-final1

Humans are fantastic at finding holes in incentive systems. Give incentives to retail employees to push an add-on product – and they’ll find a way to do so without increasing your bottom line profits. Incentivize efficient behavior for testers – finding bugs! – and they’ll open millions of stupid, mindless little defects that waste your time.

Joel Spolsky says it best:

I’m always on the lookout for these incentive schemes gone wrong. There’s a great book on the subject by Harvard Business School professor Robert Austin — Measuring and Managing Performance in Organizations. The book’s central thesis is fairly simple: When you try to measure people’s performance, you have to take into account how they are going to react. Inevitably, people will figure out how to get the number you want at the expense of what you are not measuring, including things you can’t measure, such as morale and customer goodwill.

This has far-reaching implications into motivating employees that are far from commission-based. Let’s take two simple premises that many large companies follow:

  1. Reward employees on a curve (ex. 20% great, 60% average, 20% under-performing). Force that curve at every level of the organization to make sure that no manager can claim that his team is “all stars” and thus pull the blanket of rewards to her group.
  2. Tie financial rewards for an employee are to their individual success only – that is, their relative performance compared to their peers, as viewed by a panel of superiors. The logic goes, if you tie employee rewards to product success, you won’t have stars joining struggling teams as they don’t want to be weighed down by failure.

Let’s think about what you get in the end from this explosive combination from a purely logical, reasonable employee. They observe that their own upside is completely disconnected from the success of their product. As a result, two kinds of behaviors start:

  1. Backstabbing and political play. This one is kind of obvious and everyone talks about it: if I’m rewarded based upon subjective opinions of my superiors, I better kiss up to the superiors and make my peers look bad.
  2. Poor hiring. Who are the people doing the interviewing for new talent in your organization? The same people that you apply the incentives to. “If I hire mediocre people onto my team,” the thinking goes, “I will be a shining star at the next performance review!” Therefore, I’ll give a Hire to an intentionally bad candidate! This will ensure my continuous success in the organization, while decreasing the chances of the group as a whole to succeed.

OK, you’re in shock. You’re about to say that:

  • I have to work with these mediocre people that I just brought into my group. Yeah, so what? I can hire really nice and really unproductive co-workers. I’ve seen plenty in my career.
  • If everyone does it, the company will go under and there won’t be any cash wins for anyone! Yeah, except that when I see others doing it, I don’t want to be the one left out.
  • This is plain immoral and intentionally malicious to the organization! Nobody would do this!

Do not ever place people in a situation where rational, logical behavior conflicts with their morals. Your counterpart will either suffer moral trauma (my wife really wanted to go to Hawaii, and there I am, not doing all it takes…) or do the thing that’s wrong for the business.

Here’s an intuitive way to approach this: if your friend lent you a MILLION dollars in cash tomorrow, with no contract or witnesses, with no trace of the transaction, would you be tempted to cheat them and pretend they never gave you the money? How about a BILLION dollars? An amount that will be enough to take care of EVERYONE you love and their grand-children, for the rest of their lives?..

Stop lying to yourself. There is a price for everything. People are rational beings, and they place value on things that are difficult to measure. You probably wouldn’t screw a friend for a thousand dollars, right? But for a BILLION? You would. Everyone would. Don’t put your friends in a position where their morals conflict with what’s good for them in terms of utility. And don’t ever do this to business associates, because I guarantee you – their loyalty to your business is lower than your loyalty to your friends.

Good fences make good neighbors. Good contracts make for fabulous partnerships.