The 30-minute research cap

The 30-minute research cap is a simple rule: any research task — a tool comparison, a supplier decision, a "how do others do this" question — gets 30 minutes, after which you must either decide, buy, or write down the one specific question that genuinely blocks you. Uncapped research is procrastination with a clear conscience, and it consumes more operator time than any other disguised delay I know.

Why does research expand to fill any container?

Because it always feels productive and never carries a deadline of its own. Reading a fourth comparison article produces the same sensation as progress while deferring the risk of being wrong. The information environment makes this worse: on any business question there is effectively infinite material, most of it repeating the same three points, so "I'll just read a bit more" has no natural stopping point.

There is also a status trap. Being well-researched feels like diligence, and stopping early feels reckless. But diligence is measured by decision quality over time, not by tabs opened. Most decisions an operator faces are reversible, and for reversible decisions the cost of deciding slightly early is small while the cost of deciding late compounds weekly.

This is why the cap sits inside my personal operating system as a standing rule rather than a judgement call. The whole point is that no judgement is required in the moment.

Why 30 minutes specifically?

It is not sacred; it is calibrated. For the typical operator decision — a £50/month tool, a process question, a template — 30 minutes is genuinely enough to find the consensus view, the main dissent, and the price. What it is not enough for is false comfort, which is precisely the feature. The cap forces you to notice that more reading was never going to remove the residual uncertainty. Action removes it.

Bigger decisions get bigger caps, decided in advance: a half-day for anything contractual or hard to reverse, more for hiring. The rule is not "30 minutes for everything". The rule is "every research task gets a cap before it starts, and the default is 30 minutes unless you can say why not".

How does the cap work in practice?

The mechanism, step by step. When a research question comes up, then it gets written down as a question, not held in your head — "which cold-email tool" becomes "which tool sends 25–40 emails a day per inbox without breaking deliverability". When the timer starts, then you work only from that question; anything interesting but off-question gets captured for later, not followed. When the 30 minutes end, then you must produce one of three outputs: a decision, a purchase, or the single named question that blocks you. When the output is a blocking question, then it goes to a person who knows — a supplier, a peer, a specialist — rather than back into the search bar. When the answer comes back, then you decide without reopening the research.

The discipline is in the third output. "I need to think about it more" is not a permitted result. A named question is.

What about decisions that genuinely need depth?

They exist, and the cap handles them by making depth a deliberate purchase rather than a drift. If 30 minutes ends and you can articulate why this decision deserves four more hours — irreversibility, large money, regulatory exposure — then you book those hours consciously. What you may not do is let a small decision quietly become a research project. The cap is the tripwire that makes the escalation visible.

Recruiters gave me the cleanest illustration of the alternative. In recruitment outbound, the firms that stall are rarely short of information — they have usually researched tools and data sources for months. The firms that build pipeline picked adequate tools quickly and spent the saved weeks actually sending. The research delta was near zero; the action delta was the whole result.

How do you know the cap is working?

Track it like anything else you claim to care about. Two of my personal KPIs touch this directly: decisions closed per week, and average age of open decisions. When the cap is being respected, decision age falls fast and — the part people doubt — decision quality holds. In a year of running this I can name decisions I got slightly wrong and corrected in a week. I cannot name one where the missing ingredient was a fourth hour of reading.

Ship the decision the way you would ship the work at 80%: early, consciously, and with a plan to correct on contact with reality. Research is an input. It was never the work.

---

Keep the thinking coming: The Format — a fortnightly letter on business and personal systems. Subscribe by email.