The crux: name the challenge that makes it hard.

From Good Strategy/Bad Strategy and the 2022 book that bears its name: the crux is the point where difficulty concentrates, and where effort, if it lands anywhere, must land. Martin's practice converges on the same discipline from a different direction: a strategy document that never states its crux is either hiding the difficulty or has not found it, and both are disqualifying.

Finding it

Two reliable routes. Run What Would Have to Be True across the strategy's conditions, and the crux is the pivotal condition you are least confident in. Or ask the elimination question: if we solved everything else but not this, would the strategy still fail? If yes, you have found it. Both routes work because the crux is structural, not emotional: it is not the challenge people complain about most, it is the one the logic actually turns on.

What naming it buys

The crux disciplines effort. Capability building concentrates where the strategy can actually break rather than spreading evenly across a plan, and testing goes where failure would be informative. Naming it also creates shared understanding and shared urgency, which is why Operation CRUSH worked as a memo where a goal card would have failed: everyone at Intel knew exactly what the fight was. Comfortable strategies have no crux because they attempt nothing that could fail, which is the tell of Bad Strategy and Playing to Play alike. Within the kernel, an honest diagnosis names the crux and the guiding policy is the approach to it.

How this connects to OKRs

The crux is the riskiest-assumption discipline at strategy altitude, and it should aim the quarter: the best use of an OKR is to attack the crux, isolated and made attackable, rather than to distribute ambition evenly across departments. Rumelt's own goal-setting rule, proximate objectives, follows directly: the best proximate objective is often the crux itself.