Human Performance & Leadership
Stop Letting “Speed‑First” Decisions Drain Your Team
Amy Edmondson's best-managed hospital units reported the most errors, and that finding should govern how fast you demand answers.
2026-09-171 min read
A demand for an immediate answer is rarely heard as a question about timing. It is heard as a statement about what kind of answer is acceptable, and the fastest available answer is always the confident one — hedges, caveats and "I would want to check that" all cost sentences the room has just made clear it will not pay for. What time pressure compresses is not the thinking. It is the reporting of doubt.
Amy Edmondson found this by accident in the mid-1990s while studying medication errors across hospital nursing units. She expected the better-managed units to show fewer errors. They showed more, and the interviews explained why: those units were not making additional mistakes, they were willing to write them down. The error rate in her data was measuring candour, and the units led by people who reacted badly to bad news looked excellent on paper right up until something surfaced that could not be hidden.
A leader asking for a go/no-go in five minutes is running the same experiment without meaning to. The engineer holding a half-formed worry has three options: state it and look unprepared, state it vaguely and be ignored, or say "looks fine" and be efficient. Two of the three produce a clean meeting. What degrades is the organisation's information about its own risks, and it degrades precisely in the situations where speed felt most necessary.
None of this is an argument for slow decisions. Most decisions should be fast, because most are cheap to reverse and the deliberation costs more than the error would. The distinction that matters is whether the decision can be walked back, and whether the person being rushed is the one holding information nobody else has.
Key insights
Time pressure compresses the expression of doubt long before it compresses the thinking.
Units whose leaders punish bad news produce clean reports, not clean operations — Edmondson's better-managed teams logged more errors, not fewer.
Reversibility, not importance, is the right test for how fast a decision can safely be made.
Why it matters
Speed pressure does not relent when a decision is genuinely urgent — it strips out the caveats, and the caveats are the part only the person in front of you can supply.
A team that has learned not to voice doubt looks better in every meeting and every report until the first failure that cannot be smoothed over.
Use this tomorrow
1In your next decision meeting, write down how many people voice an objection or an uncertainty; if that count is zero or one in a group of six or more, ask the two quietest people directly before you close the item.
2List the three decisions you made fastest last month and mark each one reversible or not; put any non-reversible decision that took under ten minutes back on an agenda this week.
Go deeper
Edmondson's work became the foundation of the research literature on psychological safety — the shared belief that a team is safe for interpersonal risk-taking, which she defined in a 1999 paper in Administrative Science Quarterly. The construct has since been measured across a very large number of teams and is among the more reliably replicated findings in organisational behaviour, which is not a phrase that applies to much of that field. The practical content is narrower than the popular version suggests: psychological safety is not comfort, and it does not mean lowering performance standards. It predicts learning behaviour — asking questions, reporting errors, seeking feedback — and it is the learning behaviour, not the mood, that shows up later in outcomes.
The counterweight is that deliberation has a real and often underrated cost, and leaders who read this literature sometimes overcorrect into meetings that never end. Amazon's shareholder letters offer a useful discipline here: Jeff Bezos separated one-way doors from two-way doors, arguing that decisions which cannot be walked back deserve a slow process while reversible ones should be made quickly by whoever is nearest, since the cost of being wrong is a single reversal. Most organisational pathology is applying the wrong process to the wrong door — committee review for the reversible, five-minute instinct for the permanent. The speed question is downstream of a classification the room has usually never made.
