When You’re Stuck, Think Less#
Trap one: thinking harder#
Working memory is a workbench, and everyone’s is small. When a problem has more moving parts than fit on the bench, adding effort doesn’t add capacity — it just adds fatigue, and clarity drops the longer you push.
The way out is borrowed from software: separate the concerns. Resolve two or three factors, park the rest, then bring in the next piece. You’re not solving less of the problem, you’re refusing to hold all of it simultaneously.
The spiral this avoids is familiar. Optimising A makes B worse. Fixing B costs you C. You go to bed on it, re-run the simulation at 2am, and wake up with less clarity than you started with because you’ve been accruing tiredness rather than insight.
Trap two: treating overwhelm as a capacity problem#
Overwhelm feels like evidence that there’s too much for you. It’s better read as a signal to triage.
The question that does the work: what single constraint, if removed, would make everything else simpler? You don’t need to see how the whole thing connects to know the next move.
Two tools:
- A dependency map. Dump every factor onto a page and draw which ones block which. The biggest constraint becomes visible rather than theoretical. Works for decisions, planning, prioritising, and working out what to learn next.
- A weekly hour for the non-urgent. Reserve a small block for constraints that aren’t screaming yet. The logic is sharp: if something takes six months to clear and becomes urgent in six months, it is urgent now.
The example behind this is a junior doctor on call covering 150 patients, some nights with under ten seconds of rest across five hours. Spending an extra five minutes on one procedure meant missing five calls. That forces a shift from how do I get through all of this to what actually matters right now, which is the only question that scales.
Trap three: chasing certainty#
This one is the most counterintuitive and probably the most useful.
Confidence and precision trade against each other. “Three billion, plus or minus three billion, English speakers” is 100% confident and completely useless. A drug lowering blood pressure by 10 points ± 3 is a genuinely informative claim precisely because it isn’t certain.
So when you need more confidence to act, widen the interval rather than narrowing it. Don’t hunt for the exact answer — find the range you’d bet on.
In practice: instead of nine months and forty tests chasing an exact conversion rate, conclude you’re 99% confident it lands between 2% and 7%, then design something that works across that range. Instead of either reusing last year’s message blindly or testing forever, say you’re 70% confident the old one still works and 99% confident one of five will — then test the five.
The distinction worth remembering#
Genuinely capable people reframe the problem. People who only appear capable sound confident by quietly ignoring the difficult parts.
Being honest about what you don’t know is what makes the thinking work, rather than what undermines it.
A few small things to try#
- When stuck, cut down to two or three factors and park the rest deliberately
- Draw a dependency map for something you’ve been circling
- Book one hour a week for the important thing that isn’t urgent yet
- Replace “what’s the answer” with “what range am I confident in”
