Iterate Then Commit (The Secretary Problem for Life)
Explore fast to set your bar, then go all in on the first thing that clears it
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 6
- Confidence
- 82%
Naval borrows the secretary problem from computer science and reframes it as iteration-based rather than time-based. If you will hold a role for ten years, you do not search for a set number of years; you take enough shots that by roughly a third of the way through your iterations you have seen the field and can set a bar. After that, the first candidate who meets or exceeds that bar is good enough. The counterintuitive move is that this rewards taking opportunities quickly and bailing out quickly, so you maximize the number of error-correcting iterations. He pairs it with a reframe of mastery: not 10,000 hours but roughly 10,000 iterations, where iteration means modifying with a learning and trying again, which is error correction, not mere repetition. Applied to relationships, jobs, and location, it means explore cheaply, then when you find the match, go all in and compound.
Origin
Naval recounts the secretary problem as a computer-science professor's optimal-stopping puzzle, then generalizes it to dating, jobs, and careers, and fuses it with his 10,000-iterations reframe of Gladwell's 10,000-hour rule.
Core principles
- 01Optimize on iterations, not elapsed time
- 02Bail out of losers quickly and cheaply
- 03The bar is set by how many candidates you have seen, not how long you looked
- 04Once you find the match, move your chips to the center of the table
- 05Iteration is error correction, not repetition
How to run it
- 1
Frame the decision as iterations
Decide roughly how many attempts, dates, interviews, or ventures you can realistically take at this class of decision. Iterations, not the calendar, drive the learning curve.
Watch out Elapsed time is a false clock; a year of one unchanging situation is a single iteration, not many.
- 2
Explore the first third without committing
Sample widely and cheaply through about the first third of your iterations purely to see what is out there and calibrate.
- 3
Set the bar from what you have seen
Take the best candidate you have encountered so far and make that your explicit standard for good enough.
- 4
Commit to the first option that clears the bar
After the exploration phase, accept the first candidate who meets or exceeds your bar rather than holding out for a mythical perfect option.
- 5
Bail out of losers immediately
The moment you know something will not work, leave. Staying past that point is the most common and most expensive regret.
Pro tip In failed relationships the biggest regret is almost always staying after you knew it was over.
- 6
Once matched, go all in
When you find the one relationship or business worth compounding into, move your chips to the center and commit fully; the barbell is explore-cheap then bet-big.
Watch out Being stuck in the gray middle, half-in and unsure, is the worst position in a barbell strategy.
In the wild
Naval argues modern society is far more forgiving of failure than a tribal system with twenty possible mates. You run a search function: if a date fails there are infinite more people. It is fine to have fifty failed dates in between, because once you find the mate you compound in that relationship for the rest of your life. The same logic covers fifty small failed ventures before the one business you plow into.
→ A high-variance search that ends in one compounding, all-in commitment.
Naval reframes Gladwell's mastery rule: it is not 10,000 hours but roughly 10,000 iterations, and iteration is not repetition. Repetition is doing the same thing over and over; iteration is modifying it with a learning and doing another version. If you get 10,000 error corrections in anything, you become an expert, which is why fast, decisive iteration beats mindless practice.
→ Faster mastery by maximizing error-correcting loops rather than raw hours.
Common mistakes
Committing before the bar is set
Marrying the first person you date or taking the first job without exploring means you never calibrated what good actually looks like.
Staying past the point you knew
Refusing to bail out of a clearly-doomed relationship or venture burns the iterations you need for the real match.
Confusing repetition with iteration
Doing the same thing 10,000 times without a learning attached produces no error correction and no mastery.
Is it for you?
Best for
Big, hard-to-reverse choices in dating, hiring, careers, and location where you want a principled stop rule.
Not ideal for
Low-stakes reversible choices where any deliberation is wasted overhead.
From the transcript
“about a third of the way through, you take the best person you've worked with and try to find someone that good or better.”
“It's iteration based. The number of candidates, the number of shots you took on goal.”
“I would say it's actually 10,000 iterations to mastery... iteration is not repetition.”
From the episode
44 Harsh Truths About Human Nature - Naval Ravikant - #922
Naval Ravikant