The moment you turn a measurement into a target people are paid to hit — what happens to the thing you were actually trying to fix?
Let me tell you about a city with a snake problem — and a solution so sensible it’s almost funny how badly it backfired. Inside this small, strange story is one of the most important ideas in the whole chapter, the one that decides whether the numbers a country chases actually help it.
Let me tell you about colonial-era Delhi. The city had too many venomous cobras, and the government wanted the number down. So it did the obvious, reasonable thing: it offered a bounty — a cash reward for every dead cobra you brought in. Clean, simple, measurable. Bring a snake, get paid. At first it worked exactly as intended, and the dead cobras came in.
Then people noticed something. If a dead cobra is worth money, a cobra is an asset. So enterprising residents began quietly breeding cobras — not to reduce the snakes, but to farm the reward. When the government realised what was happening and scrapped the bounty, the breeders had no more use for their now-worthless snakes, so they let them go. The city ended up with more cobras than before it ever tried to help.
The bounty measured “dead cobras handed in.” But the goal was “fewer cobras in the city.” As long as nobody was gaming it, those two things moved together and the measure looked perfect. The instant there was money in it, they came apart — and people optimised the measure they were paid for, not the goal it was standing in for.
That is the trap, and it is everywhere. Reward a factory for the number of nails it makes and you get a mountain of tiny useless nails; switch to rewarding weight and you get a few giant ones nobody can use. The measure, once it becomes the target, stops being an honest picture of the thing you cared about. For a country choosing how to size up its own progress, that is not a cute anecdote — it is the whole risk. Pick the wrong number and you don’t just mis-describe the nation; you send it chasing the wrong thing.
You get what you measure. The moment a measure becomes a target, people optimise the measure instead of the goal — and the two quietly part ways. So the number a country picks to chase is never neutral. It steers everything.
It would be easy to take the wrong lesson — “measuring is dangerous, so don’t.” But the cobra story cuts both ways. During the pandemic, one country measured how many days it took to trace a COVID contact, and made that number the target. Because the measure genuinely tracked the thing that mattered — stopping the spread — chasing it made the country better at the real goal, not worse. Same tool, opposite result. A measure can drive a nation toward the right thing just as easily as toward the wrong one. Everything hangs on whether the number you chase and the outcome you want are actually the same thing.
Explain what went wrong in Delhi to someone who’s never heard of it — and pull the general rule out of it, without stating any “law” by name.
The cobra bounty backfired; the contact-tracing target worked. Same tool, opposite results — so what actually made the difference?
Pick a number your school, team, or workplace is actually judged on. How could someone hit the number while missing the point?