Human Performance & Leadership
Decision‑Blind Spot
Alfred Sloan once adjourned a meeting because everyone agreed, saying the committee needed time to develop disagreement first.
2026-09-112 min read
Peter Drucker tells the story in The Effective Executive: Sloan, running General Motors, opened a committee meeting, found unanimous agreement on an important decision, and adjourned it. His stated reason was that the group had not yet developed enough disagreement to understand what it was deciding. Read that as a diagnosis rather than a management quirk. Unanimity arriving at the start of a meeting is evidence that the deciding happened somewhere else, in a corridor, a car ride, a one-to-one the day before, and that the meeting has been convened to witness a decision rather than to make one.
The pre-meeting is not corruption. It is how organisations actually work, and it works because the informal channel is cheap, fast and safe: a senior person can float a half-formed idea to one trusted peer without paying the cost of being wrong in front of thirty. The defect is not that the conversation happens early. The defect is that it has an invisible guest list and leaves no record, so nobody outside it can tell what was traded away, which alternatives were weighed, or whose objection was never heard. A plan assembled that way cannot be argued with, only complied with, and teams that sense this stop preparing for meetings, which makes the informal channel more decisive still.
Japanese firms formalised the same channel under the name nemawashi, literally the digging around a tree's roots before it is transplanted, in which a proposal is walked to each affected party individually before it reaches the room. The formal meeting then ratifies. What separates that from a hallway coup is not the absence of pre-alignment but its coverage and its traceability: the circuit is expected to reach everyone who will have to live with the outcome, and being skipped is a visible failure rather than a private one. The useful question for a leader is never whether the real conversation happened early. It is who was on the route, and whether anyone can name the person left off it.
Key insights
Instant unanimity is evidence that a decision was made elsewhere, not evidence that it was made well.
Pre-meeting alignment is unavoidable and often useful; the defect is the undocumented guest list, not the pre-meeting.
The auditable question is who was on the route to the decision, and who was left off it.
Why it matters
A decision the team cannot reconstruct is a decision the team cannot improve, so the same unexamined premise gets reused until it fails in public.
Consensus that arrives before the discussion is the cheapest thing to mistake for alignment and the most expensive to discover was not.
Use this tomorrow
1Before your next decision meeting, list every person who will have to change what they do if the decision passes, then count how many you have spoken to individually; hold the meeting when that count reaches all of them.
2At the end of your next meeting that reached agreement in under ten minutes, have each attendee write their strongest objection on a card, collect the cards, and count how many objections nobody had said aloud.
Go deeper
Drucker's account of Sloan is the most-quoted version, but the underlying failure has a name in group-decision research: the hidden-profile problem. In experiments where each participant is given a different fragment of the evidence, groups reliably converge on the option supported by information everyone already shares, and fail to surface the fragments held by only one person. Garold Stasser and William Titus documented this in the 1980s, and the effect is strongest when a group reaches agreement early and comfortably. A pre-aligned meeting is a hidden-profile experiment run at speed: the shared information is whatever the corridor conversation established, and the unshared fragment is whatever the person who was not on the route happens to know.
The counter-argument deserves a hearing. Forcing every decision into the formal forum is how organisations acquire the meeting load that made the hallway attractive in the first place, and nemawashi is not free either: it is slow by design, trading decision speed for implementation speed on the theory that a decision nobody resisted executes faster than one that was merely announced. The real choice is about where you want the delay to sit, before the decision or after it. Reversibility is the practical test, because for a decision you can undo inside a week the corridor is fine and the ceremony is waste.
