Jeff Bezos High-Quality Decisions: What He Learned from Warren Buffett
In September 2018 at the Economic Club of Washington, Jeff Bezos explained his decision-making philosophy: a senior executive gets paid "to make a small number of high-quality decisions"; for him, about three good decisions a day is enough; and "Warren Buffett says he's good if he makes three good decisions a year." Bezos protects decision quality through deliberate practices: scheduling "high-IQ meetings" for 10am, refusing major decisions after 5pm, prioritizing 8 hours of sleep, and using Amazon's 6-pager memo system to force structured reasoning. On gut decisions, Bezos said: "All of my best decisions have been made with heart, intuition, guts—not analysis." This post's reading: his gut is trained by preparation and written reasoning—major Amazon proposals are argued out in narrative memos, creating institutional memory that can feed future intuition. The loop, as we synthesize it: preparation → gut decision → documentation → future preparation. Bezos operationalized Warren Buffett's philosophy through Amazon's systems—the 6-pager, silent reading, Type 1/Type 2 decision framework, disagree and commit. The principle: fewer decisions, higher quality, documented reasoning.
In 2018, Jeff Bezos revealed his decision-making philosophy: "You get paid to make a small number of high-quality decisions." He credits Warren Buffett's "3 decisions a year" approach. But Bezos added something crucial: a system to capture why decisions were made—so documentation becomes tomorrow's preparation.
- 3 decisions a day — protect cognitive capacity for what matters
- 10am high-IQ meetings — schedule decisions when you're sharpest
- Gut + Documentation — intuition works because it's trained and recorded
- The flywheel (our synthesis) — documentation → preparation → trusted gut → documentation
Jeff Bezos at the Economic Club of Washington, September 2018
The Interview That Changed How CEOs Think
September 13, 2018. The Washington Hilton. Jeff Bezos—then the world's richest person—sat down with David Rubenstein for a 70-minute interview at the Economic Club of Washington.
When asked about decision-making, Bezos delivered an answer that would become one of the most-cited pieces of executive wisdom of the decade:
"As a senior executive, what do you really get paid to do? You get paid to make a small number of high-quality decisions."
"Your job is not to make thousands of decisions every day."
"If I make, like, three good decisions a day, that's enough. And they should be as high quality as I can make them."
Then came the reference that connected Amazon's culture to its philosophical roots:
"Warren Buffett says he's good if he makes three good decisions a year. So, you know, I really believe that."
Three decisions a day for a CEO whose company had just briefly passed $1 trillion in market value. Three decisions a year for the investor who built Berkshire Hathaway. Different scales, same principle: quality over quantity.
The 10am Rule
Bezos doesn't just limit the number of decisions. He's obsessive about when he makes them.
"I do my high-IQ meetings before lunch. Like, anything that's going to be really mentally challenging, that's a 10 o'clock meeting."
"And by 5 p.m., I'm like, 'I can't think about that today. Let's try this again tomorrow at 10 a.m.'"
This isn't laziness. It's protection. Bezos knows that a decision made at 6pm, when you're tired and want to go home, often isn't actually faster—because you'll second-guess it tomorrow and potentially redo it.
Bezos's Daily Architecture
The Sleep Non-Negotiable
Bezos is famously protective of his sleep—and frames it explicitly as a decision-quality strategy:
"I get eight hours of sleep. I prioritize it."
In paraphrase: if he slept four hours a night instead of eight, he would gain about 33 percent more time to make decisions—133 decisions instead of 100. But is that really worth it if the quality of those decisions might be lower because you're tired or grouchy? His answer: probably not.
The math is devastating for hustle culture: 33% more decisions isn't worth anything if they're worse decisions. At Bezos's scale, a single bad decision could cost billions.
The Hidden Cost of "More Hours"
Bezos's point: we celebrate executives who work 18-hour days, but we don't audit whether their extra decisions were good ones. At the CEO level, one bad decision made at midnight can wipe out the value of a hundred small decisions made while exhausted.
The "Gut Decision" Paradox
Here's where Bezos says something that seems to contradict everything else:
"All of my best decisions in business and in life have been made with heart, intuition, guts... not analysis."
Wait—the CEO of the most data-driven company on Earth says his best decisions aren't driven by analysis?
Here's what people miss: Bezos isn't saying "don't analyze." He's saying analysis is table stakes. After you've done the analysis, the final call—for the truly important decisions—comes from trained intuition.
And that intuition isn't magic. It's trained by:
Documentation
Major proposals at Amazon are written up as six-page narrative memos that make the case for WHY, not just WHAT. A good memo sets out the reasoning and has to anticipate the alternatives and objections the room will raise. That memo is the 6-pager, and it has its own two-sided story: <a href="/blog/amazon-meeting-rules">why Amazon banned slides for it</a>, and <a href="/blog/amazon-six-pager-problems">where the 2004 format now falls short</a>.
Pattern Recognition
When you've reviewed hundreds of decision memos—successes and failures—your brain builds models. "Gut feeling" is pattern matching against that library.
The Real Insight
Our reading: Bezos's gut decisions aren't untrained instincts. They're the output of a system that documents reasoning, preserves institutional memory, and trains intuition through structured reflection. His "gut" had been reading 6-pagers since 2004.
The Decision Flywheel
Put Bezos's habits next to Buffett's and a loop appears. This is our synthesis, not a model either man has set out—but it explains why both treat decisions as more than isolated events:
The Bezos-Buffett Decision Loop
The loop closes: Documentation becomes preparation. Preparation trains intuition. Intuition makes decisions. Decisions get documented.
It is the logic behind writing reasoning down at all: a shareholder letter or a six-pager can be reread long after the decision it argued for. Documentation isn't bureaucracy—it's how you train organizational intuition.
Type 1 vs Type 2: The Speed Framework
Bezos doesn't apply the same rigor to every decision. He distinguishes between two types:
Type 1: One-Way Doors
Irreversible decisions. High stakes. Hard to undo.
- • Acquisitions, major pivots, key hires
- • Deserve deep analysis and documentation
- • Worth the 6-pager investment
Type 2: Two-Way Doors
Reversible decisions. Lower stakes. Easy to course-correct.
- • Most day-to-day decisions
- • Decide with ~70% of desired information
- • "Disagree and commit" if needed
"Most decisions should probably be made with somewhere around 70% of the information you wish you had. If you wait for 90%, in most cases, you're probably being slow."
The failure mode Bezos warns against: treating Type 2 decisions like Type 1. When organizations apply heavy process to every decision, they become slow without becoming better.
Disagree and Commit
What happens when the team can't reach consensus? Bezos's answer: disagree and commit.
"I disagree and commit all the time. We recently greenlit a particular Amazon Studios original. I told the team my view: debatable whether it would be interesting enough, complicated to produce, the business terms aren't that good, and we have lots of other opportunities."
"They had a completely different opinion and wanted to go ahead. I wrote back right away with 'I disagree and commit and hope it becomes the most watched thing we've ever made.'"
The lesson we draw: write the disagreement down. If the reasoning on both sides is preserved and the decision proves wrong, there's a record to learn from. If it proves right, there's a record of what the dissent looked like and why it didn't win. Amazon is not alone in formalising this — McKinsey and Bridgewater built their own versions, compared side by side in <a href="/blog/structured-disagreement-framework">the disagreement protocols of three organisations</a>.
Why This Matters
"Disagree and commit" isn't about suppressing dissent. It's about preserving dissent while maintaining speed. The disagreement is on record. The reasoning is documented. The organization moves forward. And when it's time to evaluate the decision, all the context is there.
From Buffett to Bezos to Your Organization
Warren Buffett built Berkshire Hathaway on a handful of truly good decisions, explained in shareholder letters. Jeff Bezos built Amazon—which first touched $1 trillion in market value days before the 2018 interview—aiming for 3 good decisions a day, argued out in 6-pagers.
The scale is different. The principle is identical:
Protect your high-quality decisions
Don't dilute cognitive capacity across hundreds of small choices.
Schedule decisions for peak clarity
Tired decisions aren't faster—they're just worse.
Document the reasoning, not just the conclusion
Today's documentation becomes tomorrow's preparation.
Trust trained intuition—but train it first
Gut decisions work when the gut has been fed structured preparation.
Buffett trains his intuition by reading 500 pages a day. Bezos built a system—6-pagers, silent reading, disagree and commit—that trains organizational intuition at scale.
The question for your organization: What system do you have to capture reasoning, preserve institutional memory, and turn today's decisions into tomorrow's preparation?
Frequently Asked Questions
What did Jeff Bezos say about high-quality decisions?
In a 2018 interview at the Economic Club of Washington, Bezos said: 'As a senior executive, what do you really get paid to do? You get paid to make a small number of high-quality decisions. Your job is not to make thousands of decisions every day. If I make, like, three good decisions a day, that's enough. And they should be as high quality as I can make them.' He then referenced Warren Buffett: 'Warren Buffett says he's good if he makes three good decisions a year.'
Why does Jeff Bezos schedule important decisions for 10am?
Bezos explained: 'I do my high-IQ meetings before lunch. Like, anything that's going to be really mentally challenging, that's a 10 o'clock meeting.' He deliberately avoids major decisions after 5pm, saying: 'By 5 p.m., I'm like, I can't think about that today. Let's try this again tomorrow at 10 a.m.' He prioritizes decision quality over decision quantity, recognizing that tired or rushed decisions often need to be revisited anyway.
What does Bezos mean by 'gut decisions'?
Bezos said: 'All of my best decisions in business and in life have been made with heart, intuition, guts... not analysis.' But this doesn't mean he avoids analysis. It means that for his most important decisions, after all analysis is done, the final call comes from trained intuition. On this post's reading, that intuition is fed by years of preparation and by Amazon's written-memo culture. Gut decisions without preparation are guesses; gut decisions with preparation are pattern recognition.
How does Bezos think about sleep and decision-making?
Bezos prioritizes 8 hours of sleep because it protects decision quality. He said: 'If you shortchange your sleep, you might get a couple of extra productive hours, but that productivity might be an illusion because you're going to make lower quality decisions.' He rejects the macho culture of working on minimal sleep: at the 2018 Economic Club of Washington he reasoned that sleeping four hours would buy about 33% more time for decisions, and asked whether that is worth it if the quality of those decisions might be lower because you're tired or grouchy. His answer: probably not.
What is Bezos's '70% information' rule for decisions?
Bezos believes most decisions should be made with about 70% of the information you wish you had. Waiting for 90% certainty means you're too slow. He distinguishes between Type 1 decisions (irreversible 'one-way doors' that deserve deep analysis) and Type 2 decisions (reversible 'two-way doors' that should be made quickly). Most decisions are Type 2, but organizations often treat them as Type 1, creating unnecessary slowness.
What is 'disagree and commit' at Amazon?
Disagree and commit is Bezos's principle for avoiding decision paralysis: once a decision is made, everyone commits to executing it fully, even those who disagreed. Bezos has said he uses this himself: 'I disagree and commit all the time.' Our addition: if the disagreement is written down, the organization can learn from it when a decision proves wrong. It's not about suppressing dissent; it's about moving forward while preserving the record.
How did Bezos operationalize Buffett's philosophy at Amazon?
Bezos credits Buffett's 'three good decisions a year' philosophy as an influence. He operationalized it at Amazon through: (1) The 6-pager memo system—forcing structured reasoning before meetings; (2) Silent reading at the start of meetings—ensuring everyone engages with the full argument; (3) Type 1/Type 2 decision framework—knowing which decisions deserve deep analysis; (4) Disagree and commit—preserving dissent while maintaining speed. Amazon's systems capture reasoning, not just conclusions.
Argumentree Team
Executive Leadership
The Argumentree team is building the collaborative decision-making platform Argumentree. Our mission is to transform how organizations make, document, and learn from decisions.
How Argumentree Helps You Apply This
Bezos built the 6-pager. Buffett writes shareholder letters. Both methods share one goal: capture reasoning, not just conclusions. Argumentree gives your team the same discipline—automatically.
Make every decision count.
Bezos makes 3 high-quality decisions a day. Your organization can too—with structured reasoning that compounds over time. Documentation becomes preparation. Preparation enables trusted decisions.
Start Free 14-Day Trial →