Drive
Autonomy, mastery, purpose
Description
In 1945, a psychologist named Karl Duncker set a small puzzle on a table: a candle, a box of thumbtacks, a book of matches. The task was to fix the candle to the wall so it could burn without dripping wax on the floor. Most people fumbled for a while before the answer clicked — empty the box, tack the box to the wall, stand the candle in it. The trick is seeing the box not as a container for tacks but as a shelf. Two decades later, another researcher, Sam Glucksberg, ran the same puzzle with a twist. He timed people, and he offered one group cash for speed. The faster you solved it, the more you earned.
The result is the kind of thing that should not happen if we understand how motivation works. The group offered money took longer — on average, about three and a half minutes longer. The reward, meant to sharpen performance, dulled it. Change the puzzle so the answer is obvious (tacks out of the box from the start), and the incentive helps again. But for the version that needs a flash of insight, paying people made them worse. This is the small, stubborn fact that sits at the center of Daniel Pink's 2009 book Drive — a fact that the way we run offices, classrooms and sales floors mostly ignores.
Pink's argument is not that money doesn't matter. It is that the model of human motivation most organizations still run on — carrots for good behavior, sticks for bad — was built for a kind of work that increasingly isn't the work we do. Somewhere between the assembly line and the knowledge economy, the operating system quietly stopped fitting the machine.
The question we’re asking : If the rewards we trust most can quietly make certain work worse, what actually moves people to do it well?What we’ll see : How a stack of experiments overturned the carrot-and-stick model, and the three drives Pink says replaced it.
Table of contents
01 Chapter 1 — The candle problem and the trouble with rewards
Pink opens Drive with a quiet admission from behavioral science: the rewards we lean on hardest often backfire. The candle problem is his opening exhibit, but it is far from alone. For roughly half a century, researchers kept running into the same inconvenient pattern, and kept being surprised by it, because it contradicted the assumption baked into almost every workplace — that if you want more of a behavior, you pay for it, and if you want less, you penalize it.
The person Pink credits most is Edward Deci, who ran a now-famous experiment at Carnegie Mellon around 1969 using a Soma puzzle, a set of blocks people genuinely enjoyed assembling. One group got paid to solve puzzles; the other didn't. During a free break, when the experimenter left the room, the unpaid group kept playing for the fun of it. The paid group, once the payments stopped, lost interest faster. Deci's conclusion was uncomfortable: paying people for something they already liked doing could turn play into work, and drain the pleasure that had powered it.
02 Chapter 2 — What a reward actually does to a task
To make sense of why incentives help some work and hurt other work, Pink borrows a distinction that runs through the research: between algorithmic tasks and heuristic ones. An algorithmic task has a known path — follow the steps, reach the result. A heuristic task has no set path; you have to experiment, improvise, invent the route as you go. Stuffing envelopes is algorithmic. Writing a strategy, designing a product, cracking the candle problem — heuristic. The if-then reward, Pink argues, is tuned almost perfectly for the first kind and quietly corrosive to the second.
The mechanism isn't mysterious. A reward puts a spotlight on the finish line, and spotlights are useful when you can see the path to the line. But when the solution lies off to the side, the spotlight becomes a blinker. Attention contracts around the obvious moves, and the sideways glance that insight depends on gets crowded out. This is why Glucksberg's cash made solvers slower on the hard candle problem but faster on the easy one. The incentive did what incentives do — it focused people — at the exact moment focus was the wrong tool.
03 Chapter 3 — Autonomy, mastery, purpose
Pink's answer is that human beings have a third drive, beyond the biological one and the reward-and-punishment one — an intrinsic drive to direct our own lives, to get better at things that matter, and to be part of something larger than ourselves. He calls the old operating system Motivation 2.0 and this one Motivation 3.0, and he names its three elements plainly: autonomy, mastery, purpose. The whole second half of the book is really an elaboration of those three words.
Autonomy is the desire to be self-directed, and Pink breaks it into four parts: control over your task, your time, your technique, and your team. His showcase example is the company Atlassian, which gave engineers occasional free days — "FedEx Days," so named because you had to deliver something overnight — to work on anything they wanted, however they wanted. The results, in new fixes and ideas, outran what the ordinary schedule produced. Management's instinct, Pink argues, is control; what the work actually rewards, more often, is letting people steer.
04 Chapter 4 — Where the carrot still earns its keep
It would be easy to read Drive as a flat verdict — rewards bad, intrinsic motivation good — and Pink is at pains not to be read that way. The more careful claim is about terrain. Carrots and sticks are not broken; they are specialized. They work well, even now, for routine, algorithmic tasks where the path is clear and the work isn't especially interesting. For those jobs, Pink says, a well-designed if-then reward can be exactly right, and pretending otherwise would be its own kind of naivety.
He even sketches how to pay people on creative work without poisoning it. The first move is to get compensation right and then off the table: pay fairly, pay a bit above the norm, so money stops being a source of anxiety and recedes as an issue. After that, the useful rewards tend to be the unexpected ones — the "now that" reward offered after the fact, rather than the "if-then" reward dangled before it. A surprise bonus for finished work doesn't reframe the task in advance, so it doesn't drain the task's own pull. Praise and feedback, specific rather than generic, do similar work for free.
05 Conclusion
Set a candle, a box of tacks and some matches on the table, and whether a cash prize helps or hurts depends entirely on how hard the puzzle is. That small reversal, running through decades of experiments that organizations mostly filed and forgot, is the whole of Pink's case in miniature. The carrot focuses, and focus is a gift when you can see the goal and a handicap when the answer is off to the side. Pink's book is an argument for noticing the difference before we pay for it.
Related books
