Compensation systems are small-scale wicked problems: They operate on groups of people, it’s difficult to run independent trials, the system changes with everything you do, unintended consequences are easy to achieve, and it’s hard to judge success. But if you’re the leader of a company you have no choice but to wrestle with this problem. One big but common mistake you should avoid is conflating compensation with motivation.

When we use compensation as motivation, we try to reward people to get them to do what’s good for the company. Daniel Pink’s book Drive is the best, simple explanation I know of for what motivates knowledge workers. In a nutshell, Pink says that mastery, autonomy and purpose are effective, intrinsic motivations. Extrinsic motivations like money, titles, and rewards have at most a short-term effect on desired behavior. If the research that Pink describes in Drive is correct, then it stands to reason that compensation is unlikely to be effective at motivating people. On the other hand, if compensation systems are perceived to be unfair, extrinsic rewards can be powerful de-motivators.

You might conclude that if extrinsic rewards like money are not effective at motivating people, then it must be that people aren’t interested in money. Yet that’s clearly not the case. Money is of course legitimately and critically important. It allows us to feed, shelter and protect ourselves and our families. It provides for luxuries, small and large. Money definitely matters to people.

If you create a reward system with money at its core, people will pay an inordinate amount of attention to the system. They will study the rules, look for loopholes, model alternative strategies, and generally spend a lot of time and creative energy figuring out how to get the most money from the system. I think it’s telling that we use the term “game the system” to describe this behavior — if you set up an interesting game, people will play it. Perhaps doing well at the game, independent of the explicit reward of the game, is motivating in and of itself. In effect, winning the game provides payoffs to individuals as a form of mastery.

If money is a poor motivator, why do so many companies rely on it for that purpose? Apparent simplicity. It appears to be simple to set up a compensation game based on money. Much simpler, for example, than creating a company culture and job roles that feed intrinsic motivation.

In theory, a perfect compensation game would be one that only rewards individuals for desirable outcomes for the company with no unintended consequences and no money spent unnecessarily. In practice, it seems to me that it’s very easy to build games that have two fundamental flaws: Either the reward can be won with little or no impact on the company, or maximizing individual winnings looks beneficial in the short-term, but actually hurts the company in the long-term. Given the complexity of even relatively simple businesses like innovation services firms, your chances of designing a perfect game seem small, and therefore the most likely outcome is time, talent and creativity wasted on outcomes of dubious or even detrimental value to the company.

I know I can’t craft a perfect compensation game. Instead, my approach is to avoid the powerful de-motivation of unfair compensation while focusing effort on creating conditions and opportunities for employees to thrive through the intrinsic motivation channels of mastery, autonomy and purpose.