The Six-Month Passion Test
Don't trust your identity about what you'll love — give the unfamiliar job six months, then judge.
- Difficulty
- Easy
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 88%
Bosworth twice told Zuckerberg no — he wasn't an ads guy, he wasn't a hardware guy — and was wrong both times. His rule: your self-concept ('I'm an AI infrastructure person') is a bad predictor of what you'll become passionate about, so give unfamiliar work six months before concluding you can't love it. Passion is not a fixed input; it's often the output of enough exposure. Pair that with aggressive rotation early in your career and treat variety as compound interest that only starts to look impressive after about ten years.
Origin
Bosworth's own arc: he rotated roughly every six months for years (anti-spam, News Feed, site speed and infrastructure, engineering bootcamp, messaging, groups), watched peers on single tracks get promoted faster, and then went 'vertical' when he took Ads and stayed five years. He frames it with the Karate Kid metaphor — painting fences and waxing cars, then discovering you know karate.
Core principles
- 01Passion is more often discovered than predicted — 'I'm a software guy, not a hardware guy' is an identity claim, not evidence.
- 02Six months is the right unit of test: long enough to get past the unfamiliarity, cheap enough to abandon.
- 03Move when you stop learning, not when you stop being comfortable.
- 04Early-career variety is compound interest — the first ten years don't look impressive, then they do.
- 05The best jobs are the ones deep enough that you never stop learning; when you find one, stay.
- 06Specializing into a domain you've already exhausted is the mid-career trap, even when it comes with promotions.
How to run it
- 1
Notice when you're saying no from identity, not evidence
When an opportunity is offered and your instinct is 'that's not who I am', flag it. Bosworth said no to ads because he had a strong self-concept as an AI infrastructure product person, and no to hardware because he was 'a software guy'. He was wrong both times.
Pro tip Ask what specifically about the work you've actually tried and disliked, versus what you're inferring from a label.
- 2
Commit six months of genuine engagement
Take the job and give yourself the space to find the interesting structure inside it. Bosworth discovered ads was a three-sided marketplace that felt like playing chess against the rest of the industry — a fascination invisible from the outside.
Pro tip Look for the hidden game: the strategic structure, the constraint, the system nobody outside the domain sees.
Watch out Six months of half-hearted presence proves nothing. The test only works if you actually engage.
- 3
Judge honestly at the checkpoint
If after six months you haven't found the passion, move on. Bosworth is explicit that there are jobs where he never found it — and that the consequence of staying is quitting, being fired, or doing bad work you don't care about.
Pro tip The failed six months is not wasted: you leave with new skills, new relationships and a broader map of the company.
- 4
Rotate on learning, not on promotion pace
Move because you're bored and not learning enough. Accept that you may appear to progress more slowly than peers on a single track — that's the cost of the compounding.
Pro tip Track a simple monthly question: am I still learning enough here to justify the seat?
Watch out Don't confuse the discomfort of a hard job with the boredom of an exhausted one. Only the second is a signal to leave.
- 5
When you find the deep well, stop rotating and go vertical
Some jobs are deep enough to keep teaching you for years. Bosworth's ads role taught him for five straight years, and that's when his career went vertical. Stay in those.
Watch out Staying past the point where the learning stops turns the deep well into the mid-career trap you rotated to avoid.
In the wild
When Zuckerberg told Bosworth to go work on ads, Bosworth said no — he had a strong identity as an AI infrastructure product person and didn't think he had a passion for it. He took it anyway and found a three-sided marketplace that felt like industry-scale chess.
→ He ran ads and business platform for five years, learned continuously, and it became the inflection point where his career went vertical. He then made the identical mistake with hardware ('I'm a software guy') and was wrong again.
For years Bosworth changed jobs roughly every six months — anti-spam, News Feed, site speed and infrastructure, engineering bootcamp, messaging, groups — while peers in his cohort stayed on single tracks and got promoted past him. He describes it as painting fences and waxing cars.
→ By the time he took Ads he had relationships and cross-domain understanding no single-track peer had. Several of those faster-promoted peers later got bored in domains they'd exhausted, with nowhere to jump.
Common mistakes
Letting your professional identity veto opportunities
'I'm not a hardware person' feels like self-knowledge but is usually just unfamiliarity wearing a costume. Bosworth got this wrong twice with the two biggest jobs of his career.
Optimizing for promotion velocity early
Single-track specialists get promoted faster in the first few years. The cost shows up mid-career when they've exhausted the domain, have no other skills, and have nowhere to jump.
Rotating out of a job that's still teaching you
The trigger to move is 'I've stopped learning', not 'it's been six months'. Bosworth stayed five years in ads precisely because it never stopped teaching him.
Is it for you?
Best for
Early- and mid-career operators deciding whether to specialize deeply or rotate across unfamiliar functions
Not ideal for
Deep technical crafts where mastery genuinely requires a decade of unbroken focus, or anyone whose role change carries a high switching cost (visa, equity cliff, sole-earner constraints)
From the transcript
“willing to learn aggressively I would move because I wasn't learning enough I was bored”
“don't be jump into new things give it six months if it's not the thing no problem like you just built a ton of new…”
“that jump that's early career optimize for learning optimize for like think about a compound interest like it's it's like the first like 10 years…”
“I do think when I find something I'm passionate about that's good what I have learned since then is to give myself the space to…”
From the episode
Making Meta
Andrew ‘Boz’ Bosworth (CTO)