n o ren
Human Performance & Leadership

Who Wins When Leaders Pause?

Amazon's executives sit in total silence for thirty minutes before every major meeting — reading, not talking.

Deliberate silence before a big decision is not dead time — it is when the brain quietly cross-checks a choice against everything it already knows. Psychologist Ap Dijksterhuis's 2006 research on complex, multi-attribute decisions found that people who took a short, distraction-filled break before choosing consistently picked better options than people who kept deliberating out loud right up to the vote. Talking a decision through in real time tends to reward whoever argues most confidently, not whoever has the best information; sitting with a problem in silence lets quieter evidence compete on equal footing.

Amazon built this insight directly into how it runs meetings. Since 2004, Jeff Bezos banned PowerPoint decks from executive reviews and replaced them with six-page narrative memos written in full sentences. Every meeting now opens with roughly thirty minutes of total silence while the room reads the memo start to finish before anyone is allowed to speak. Bezos has said the logic is simple: a bullet-point deck lets a weak idea hide behind a slick slide and a confident presenter, while a full written narrative forces the author to show their actual reasoning and forces the room to absorb the whole argument before reacting to whoever talks first.

The payoff shows up in what doesn't happen next. No one anchors the discussion on the first person to speak, because everyone starts from the same fully-read memo instead of a skimmed slide deck. Objections surface after people have actually finished thinking through the material, not in the first ninety seconds of a live pitch. The meeting gets slower at the start and faster at the end, trading a little upfront patience for a decision that doesn't have to be revisited.

A short period of silence before group discussion measurably improves decisions on complex, multi-variable choices.
Amazon's ban on PowerPoint and adoption of six-page silent-read memos is a real, still-standing practice, not a one-off experiment.

Ignoring this and letting the loudest voice open the discussion lets weak arguments win by confidence instead of by evidence.

Meetings that skip the reading period default to reacting to whoever speaks first, burying dissent that only surfaces after people have had time to actually think.

1
Before your next decision meeting, write a one-page narrative memo instead of a slide deck, and open the meeting with 10 minutes of silent reading before any discussion starts.
2
In your next status update, count how many people spoke before the most junior person in the room did; if it's more than two, build in a silent-write step next time so their view isn't anchored by others' comments.

Dijksterhuis's original studies used car and apartment choices with a dozen or more attributes, deliberately too complex to weigh consciously in the time given; the 'unconscious thought' group still outperformed the conscious deliberators when picking the objectively best option. Later replications have been mixed on the size of the effect, but the core mechanism — background processing without the interference of active reasoning — is well supported by neuroimaging showing continued relevant brain activity during a break from a problem.

The Amazon practice works partly because the memo format itself forces better thinking before the room ever sees it — a six-page narrative with no bullet points is much harder to write badly than a slide deck, because there's nowhere to hide a weak transition. The silence is as much a filter on the writer as it is a decision aid for the room.