Amazon's 'disagree and commit' turns dissent into speed

In his 2016 annual letter to shareholders, published in April 2017, Jeff Bezos argued that Day 2 means stasis, irrelevance and decline, and that one of the defenses is high-velocity decision making — yet large organizations tend to make high-quality decisions slowly, because consensus takes time and produces mediocre calls.

Let anyone with conviction ask 'will you gamble with me?': debate honestly, then disagree and commit — so disagreement never becomes a reason to stall

Bezos paired 'disagree and commit' with three supporting rules: treat reversible decisions as light-weight 'two-way doors,' decide with about 70% of the information you wish you had, and escalate true misalignment quickly instead of letting stamina decide. His own example: he doubted an Amazon Studios project, told the team, then wrote back 'I disagree and commit and hope it becomes the most watched thing we've ever made.'

The letter turned the phrase into Amazon's published Leadership Principle 'Have Backbone; Disagree and Commit,' and it became one of the most quoted decision doctrines in management writing — used by teams far beyond Amazon as the answer to meetings that circle without deciding.

In April 2017 Jeff Bezos published his 2016 annual letter to shareholders, best known for its warning about 'Day 2': stasis, followed by irrelevance, followed by excruciating decline. To stay in Day 1, he argued, a company needs customer obsession, resistance to proxies, eager adoption of external trends and — critically — high-velocity decision making. Big companies, he wrote, make high-quality decisions but make them slowly.

The letter's mechanism was a three-word phrase: 'disagree and commit.' If someone has conviction but no consensus, they ask: 'I know we disagree on this but will you gamble with me on it?' Bezos said it runs both ways — he greenlit an Amazon Studios project he doubted, then committed: 'I disagree and commit and hope it becomes the most watched thing we've ever made.' He paired the rule with two-way-door thinking (reversible decisions get light process), a 70-percent-information threshold, and quick escalation when goals genuinely misalign.

The phrase outlived the letter. Amazon codified it as the published Leadership Principle 'Have Backbone; Disagree and Commit,' and teams inside and outside the company adopted it as the antidote to consensus-driven delay. Commentators later noted the practice predates Amazon — it has been attributed to Sun Microsystems and Intel leaders — and that it fails when teams skip the hard part: actually disagreeing candidly before they commit.

It separates two things consensus confuses: honest disagreement and whole-hearted commitment.

It puts risk with the person who has conviction: they ask for a gamble instead of waiting for permission.

The 70% and two-way-door rules make being wrong cheap, so teams stop waiting for certainty that never arrives.

Quick escalation replaces the stamina contest that otherwise decides genuine misalignment.

Fast organizations separate debate from commitment: let conviction override consensus, decide on 70% of the information, and treat reversible choices as cheap enough to correct later

Amazon lists 'Have Backbone; Disagree and Commit' among the Leadership Principles its teams use every day, and the 2016 letter became a reference point in management writing about decision velocity. Later commentary added nuance: the phrase has been credited to earlier leaders such as Sun Microsystems' Scott McNealy and Intel's Andy Grove, and writers warn it breaks down when teams skip candid disagreement — commitment without argument is compliance, not conviction.