Three Questions at the Heart of Human Behavior

I have a problem. When people ask me what I do, I tell them I work in organizational development, specializing in metric and incentive alignment. Then I watch their eyes glaze over in a mix of boredom and decades of corporate trauma. The accumulated memory of every dashboard that measured the wrong thing and every bonus plan that paid someone for work they did not do should give anyone a full-body shiver.

The phrase is not inaccurate, but it is a bit soulless. I say it because it takes the shortest route to the center of my work, but it doesn’t tell the story I would like it to. I know how sharply those two words cut to the center of human behavior. It’s my job to make that clear to others in a way that doesn’t feel like jury duty.

The trouble is that the phrase names instruments instead of the thing they are for. Underneath those words there is a hidden truth about what it means to be human. There are three questions that every person answers before doing anything.

Can I do it? Will it be worth it? Is the choice even mine to make?

Can I do it?

Nobody moves toward a goal they do not believe is reachable. This sounds too obvious to be worth saying until you are responsible for getting several hundred people to move toward the same thing, at which point it becomes most of the job.

How does a person ever come to believe a goal is achievable? They gather information until they are no longer uncertain about the question. That is all a measurement is: any method of gathering information that reduces uncertainty. The most human and most basic form of measurement is to ask a question.

That definition matters, because “metrics” makes people picture a dashboard, and a dashboard is only one of the crude forms measurement can take. If something is real in any sense, it can be measured. The reverse is also worth holding onto: if something is easy to measure, you are probably not getting much value out of measuring it, because easy things tend to be the things everyone already knows. Things that are easy to measure tend to be discrete and mechanical, while the outcomes we really want to measure and describe can oftentimes be a lot more messy.

Not everything that can be measured should be measured. Every measurement costs something to take, and it costs something else that is harder to see, which is the attention it pulls off everything you are not measuring. When the cost of the measurement exceeds the value of the uncertainty it removes, you have gained nothing. I have written elsewhere about how the risk in a metric lives in the system arranged around it rather than in the number itself, and that argument becomes even more clear through this perspective.

The point for now is narrower. Measurement is how a person answers the first question. Get it wrong and people conclude the goal is out of reach, or they conclude it is within reach when it is not, and either way they act on a picture of the world you handed them. Even worse, the mere act of measurement may communicate a sense of value about the thing being measured which is not accurate, and distorts decisions rather than sharpening them.

Will it be worth it?

The second question is where incentives come in, and where most conversations about incentives go wrong immediately.

When I use the word, I mean something far larger than the narrow band people picture. Salary, bonus, and commission are real, and they are a thin slice of what moves anyone. People also work for belonging, for identity, for the chance to get better at something difficult, for the sense that what they do is consistent with who they believe they are. Self-determination theory has been mapping that territory for four decades, and it keeps landing on the same three durable drivers: autonomy, competence, and relatedness.

I am aware that a certain kind of reader hears all of this as soft. Let me approach it from the one direction that is difficult to dismiss.

My other training is in fraud examination. Researchers studying why people commit occupational fraud converged on a model, usually credited to Donald Cressey, holding that three elements must be present together before fraud is committed: motivation, opportunity, and rationalization.

Two of those map cleanly onto what I have already described. Motivation is the belief that a goal will be worth achieving, which is the second question. Opportunity is the belief that the act can be carried out without being caught, which is simply a question about whether the crime is achievable.

Rationalization is the one element worth stopping on.

It is not a footnote to the model. It is a necessary element. A person can have a powerful motive and a wide-open opportunity, and still not act, until they have found a way to square the act with who they believe they are.

Consider how this sounds from the inside. No one arrives at their desk in the morning, rubbing their hands together like a cartoon supervillain, thinking, “if I make one little change to how we bill our clients, I could make millions, and nobody would ever know!” People have an extremely difficult time thinking in terms that paint themselves as a villain. We are not built that way, and thank goodness we aren’t. What they think sounds closer to this: everyone in our industry structures billing this way, our competitors have been doing it for years, it is practically expected, and if we hold out, we are the ones disadvantaging our own people. The act does not feel like theft to the person committing it. It feels like catching up.

Now look at what the model is telling you. Criminology had to build a term for self-respect into its explanation of crime, because without it the other two elements do not account for who acts and who does not. People who are willing to steal are still not willing to think of themselves as thieves. That is not a soft observation about human nature. It is the shape of the best-established explanation we have for why people steal, and that explanation does not function without the term. Self-respect is priced into the decision alongside the money.

I find that to be genuinely good news about what it is to be human. It is also good news for anyone trying to build an organization, because it means the levers that reach people most reliably are not the expensive ones. Every successful leader knows how powerful a motivator it can be to make sure people feel good about the work they are doing and who they are doing it with.

Is the choice even mine to make?

Here I want to revise something I have said before.

The first two questions are not mine. Achievability and perceived worth are expectancy theory, which Victor Vroom set out in 1964 and which has been carrying organizational psychology ever since. What I have added is the third question, and I have been describing its role incorrectly.

I have presented this frame as three elements of one kind, as though achievability, worth, and autonomy were three boxes to be checked in sequence. I know that is not quite right. The first two are gates. The third works differently.

The tempting move is to fold autonomy into the second question and be done with it. If people value having chosen a thing, then choice is one more source of worth, and the model collapses to two clean terms. It is a tidy solution, but I think it is wrong, for a reason that can be tested rather than merely asserted.

If autonomy were only another input to the perceived worth of a goal, a large enough reward would compensate for taking someone’s choice away. Raise the payment high enough and the missing autonomy stops mattering. That is not what happens. Deci, Koestner and Ryan, reviewing decades of experiments in 1999, found that expected tangible rewards contingent on doing, finishing, or performing a task reduce people’s willingness to keep doing it when nobody is watching. Adding value to the transaction decreased the behavior. Verbal recognition, which carries no tangible or monetary value at all, increased it.

A model where worth is a single quantity cannot produce that result. Something other than the size of the payoff is being changed.

I would now describe the frame as two gates and a multiplier. Achievability and perceived worth determine whether behavior appears. Autonomy determines what kind of behavior appears, and whether it is tackled with a sense of purpose and creativity, or with the bare minimum of effort that is needed to check the box.

Skip the third question and you still get behavior. What you get is compliance. Compliance does what is measured and stops precisely there. It ends when the measuring ends. When the measure and the goal it was meant to represent come apart, which they eventually do, compliance follows the measure, because the measure is the only part of the arrangement the person feels in a meaningful way.

Clear the third question and you get ownership. Ownership does the thing nobody wrote down. Ownership notices the problem that no metric was pointed at, because the person holding it understands what the work is for and treats that understanding as theirs to act on.

What this lets a leader predict

Across years in operational roles, I watched capable, talented people decide over and over that an action that would have plainly helped the company was not worth taking. I do not think this is rare or industry-specific. I think anyone who has held a job has watched it happen and has made the same decision themselves at least once.

I have watched people stay silent about a broken step in a workflow, one they could see was exposing the company to real risk, because raising problems had never been visibly rewarded and had occasionally been punished. I have watched people send work up the chain without checking it, knowing the number attached to their name counted how much they produced and said nothing about whether it was any good. I have been in workplaces where the phrase employees repeated to each other most often was that no good deed goes unpunished.

None of those people were lazy, disengaged, or short on character. Every one of them was answering the three questions correctly, given the information their environment had supplied. The environment had told them the useful action was not worth it, or was not achievable, or was not theirs to take. They believed it, because it was true.

This is where the frame earns its keep, because the three questions do not merely describe what happened. They predict which repair will work.

A leader working from two questions has two moves available when behavior does not appear. Make the goal easier or make the reward bigger. Both are reasonable. Both are sometimes correct. When the missing element is the third question, both make the situation worse, because a larger reward attached to a goal a person had no part in choosing is a more forceful reminder that the choosing was done elsewhere.

That diagnosis is not reachable by anyone who doesn’t take all three of these variables seriously.

It also points at something larger than any single repair. When a system of metrics and incentives is built deliberately, so that people can see a goal is reachable, recognize why it is worth reaching, and feel it is theirs to reach for, intelligence and adaptability stop being things leadership has to demand. They become properties the organization produces on its own, because ownership pushes people to keep asking the questions that carry those decisions further. Organizational Intelligence does not happen merely by hiring intelligent people. It is the output of a structure that has to be built deliberately.

Back to the story

None of this is what people expect when I say metric and incentive alignment, but that’s my fault. The phrase describes my tools. It says nothing about the questions the tools exist to answer. It doesn’t talk about the conflicts leaders face or the paths that lead to resolution and growth.

So here is the plainer version, and the one I should have been using all along.

Most companies have a set of values printed on a wall. I show them how to engineer those values into the structure of the organization itself.

References

1. Cressey, D. R. (1953). Other People’s Money: A Study in the Social Psychology of Embezzlement. Free Press. Patterson Smith reprint, 1973. Cressey’s own first element was a non-shareable financial problem rather than motivation generally. The three-element model is widely attributed to him and the attribution is debated in the literature. It is an explanatory model in criminology, well established and openly criticized for a lack of rigorous empirical testing, rather than a validated experimental finding.
2. Deci, E. L., Koestner, R., and Ryan, R. M. (1999). “A Meta-Analytic Review of Experiments Examining the Effects of Extrinsic Rewards on Intrinsic Motivation.,” Psychological Bulletin, 125(6), 627 to 668.
3. Deci, E. L., and Ryan, R. M. (1985). Intrinsic Motivation and Self-Determination in Human Behavior. Plenum.
4. Vroom, V. H. (1964). Work and Motivation. Wiley.

Two claims in this article are mine rather than findings from the work above. That anything real can be measured in some form, and that the most basic form of measurement is asking a question, are positions I hold and argue from experience. The two-gates-and-a-multiplier structure is also my own, offered as a refinement to the expectancy frame rather than as a result drawn from it.

Related: A Metric’s Risk Is Not a Property of the Metric