What the Regret Minimization Framework Really Is
The regret minimization framework is a decision-making tool Jeff Bezos described in a 2001 interview with the Academy of Achievement. The idea fits in one move: project yourself forward to age 80, look back on your life, and choose the option that leaves you with the fewest regrets.
Here's how Bezos put it: "I wanted to project myself forward to age 80 and say, 'Okay, now I'm looking back on my life. I want to have minimized the number of regrets I have.'" He added the part that makes it work: "I knew that if I failed I wouldn't regret that, but I knew the one thing I might regret is not ever having tried. I knew that that would haunt me every day."
Notice what the framework quietly swaps out. Most people facing a big choice optimize for avoiding failure. Bezos optimized for avoiding regret. Those sound similar, but they point in opposite directions. Fear of failure keeps you in the safe job. Fear of regret pushes you out the door. Same person, same facts, different question, different life.
The framework is deliberately long-range. You aren't asking how you'll feel next month when the paycheck stops, or next year if the idea flops. You're asking how the 80-year-old version of you will feel about this exact fork in the road, decades after the short-term pain has faded. That time shift is the whole trick, and there's real psychology behind why it works.
The 1994 Decision That Built Amazon
In the spring of 1994, Bezos was a senior vice president at D.E. Shaw, a quantitative hedge fund in New York. He was 30, well paid, and on a clear upward track. Part of his job was scouting internet business opportunities, and that's when he ran into a number that wouldn't leave him alone: web usage, he later said, was growing at about 2,300 percent a year.
A number like that changes the math on everything. Bezos drew up a list of products that could sell well online and landed on books, partly because there were millions of titles and no physical store could stock them all. But acting on the idea meant quitting a job most people would kill for.
He told his boss, David E. Shaw. The two went for a long walk in Central Park, roughly two hours, and Shaw's response was more thoughtful than dismissive. He agreed it was a good idea, but added that it would be a better idea for someone who didn't already have a good job. Shaw asked him to think it over before doing anything.
So Bezos took about 48 hours and ran the exercise. He pictured himself at 80. Would that man regret leaving a Wall Street bonus behind? Almost certainly not. Would he regret watching the internet explode from the sidelines, never having tried? That one, Bezos decided, would haunt him. Once he framed it that way, he said, "it was incredibly easy to make the decision."
The rest is the kind of story that gets retold because it's true. Bezos and his wife MacKenzie left New York and headed west. She drove while he typed a business plan on a laptop in the passenger seat. Amazon was incorporated on July 5, 1994. The company that reshaped retail started as a bet made not on spreadsheets, but on which future self would carry the heavier regret.
Why Regret Beats Fear: The Science of Looking Back
The framework isn't just a nice story. It's built on a real and well-replicated finding about how human regret actually behaves over time.
Psychologists Thomas Gilovich and Victoria Medvec studied this directly. In their 1994 paper in the Journal of Personality and Social Psychology and their 1995 review in Psychological Review, "The experience of regret: What, when, and why," they found a clear pattern. In the short term, people tend to regret their actions, the things they did and wish they hadn't. But over the long run, the balance flips hard. Looking back across a whole life, people overwhelmingly regret their inactions, the things they didn't do.
Regrets of action have a short half-life. You did the embarrassing thing, it stung, and then you rationalized it, learned from it, and moved on. Regrets of inaction don't decay the same way. The road not taken stays open in your imagination forever, and your mind keeps filling it with everything that might have been.
That's exactly the asymmetry Bezos exploited. By projecting to age 80, he moved the decision out of the short-term zone (where the fear of a failed action looms largest) and into the long-term zone (where the ache of inaction dominates). The 80-year-old isn't scared of a bruised ego. He's haunted by the thing he never tried.
The pattern shows up outside the lab too. Bronnie Ware, a palliative carer, wrote down the most common regrets of her dying patients in her 2011 book "The Top Five Regrets of the Dying." The number one regret was this: "I wish I'd had the courage to live a life true to myself, not the life others expected of me." Her second most common was "I wish I hadn't worked so hard." The regret minimization framework is, in a sense, a way to hear that from your future self before it's too late to act on it.
How the Framework Actually Works
You can run the exercise in four steps. It takes minutes, but the honesty it requires is the hard part.
- Name the real decision. State the fork plainly. Not "should I be happier at work" but "should I leave this job to start the thing I keep thinking about." Vague inputs produce vague answers.
- Fast-forward to 80. Picture yourself old, looking back. This isn't wishful visualization. It's a deliberate time shift to escape the gravity of short-term fear, which the research says fades anyway.
- Score each path by regret, not by outcome. For each option, ask: will my 80-year-old self regret this? Crucially, you separate trying from succeeding. Bezos knew he wouldn't regret failing. He'd only regret not trying. Regret of inaction is the thing you're minimizing.
- Choose the path with the least lifelong regret, then commit. Once the regret gap is obvious, the decision usually makes itself. If it doesn't, that's useful information: the two paths may be closer than your anxiety suggested.
The framework's power is in what it removes. It strips out the noise of this quarter's numbers, other people's expectations, and the loud, temporary voice of fear. It replaces all of that with a single quiet question from the one judge whose opinion you can't dodge: the person you're going to become.
One caution about what "trying" means here. The framework rewards the attempt, not recklessness. Bezos didn't quit on a whim. He'd spotted a real trend, done the analysis, picked a specific product, and only then asked the regret question to break the tie between a safe yes-brain and a fearful gut. The exercise is the final filter, not the entire decision.
Regret Minimization vs Other Decision Frameworks
Regret minimization is one tool in a larger kit. It's powerful for a specific class of decision and clumsy for others. Here's how it compares to four other well-known models.
| Framework | Core question | Best for | Blind spot |
|---|---|---|---|
| Regret Minimization (Bezos) | What will I regret at 80? | Big, identity-shaping, once-in-a-decade choices | Small, reversible, everyday decisions |
| Expected Value | What's the probability-weighted payoff? | Repeatable bets with knowable odds | Rare, one-shot decisions where odds are guesses |
| Type 1 / Type 2 Doors (Bezos) | Is this reversible or not? | Sorting decisions by speed and care needed | Doesn't tell you what to choose, only how fast |
| 10-10-10 (Suzy Welch) | How will I feel in 10 minutes, 10 months, 10 years? | Emotional and relationship decisions | Can over-weight near-term feelings |
| Premortem | If this fails, why? | Stress-testing a plan you've mostly made | Assumes you've already chosen a direction |
Bezos himself used more than one of these. In his 2015 shareholder letter (published in April 2016) he introduced the idea of Type 1 and Type 2 decisions. Type 1 decisions are one-way doors, consequential and nearly irreversible, so you make them slowly and carefully. Type 2 decisions are two-way doors you can walk back through, so you make them fast. His warning was that big organizations often apply the slow, heavyweight Type 1 process to reversible Type 2 decisions, and grind to a halt.
That pairing matters. The one-way-door test tells you which decisions deserve the regret exercise at all. Reserve regret minimization for the true one-way doors. For two-way doors, just decide quickly and adjust. If you want to see how founders combine models like these into a repeatable system, our piece on how founders build knowledge systems traces the patterns.
Where the Framework Breaks Down
No mental model survives being used everywhere, and this one fails in predictable ways. Knowing them keeps you from misusing it.
It's the wrong tool for small decisions. Your 80-year-old self does not care which laptop you buy or whether you take the 9am or the 10am flight. Run the regret exercise on trivial choices and you'll turn ordinary life into a series of dramatic deathbed referendums. Save it for the one-way doors.
The framework can also rationalize recklessness. "I don't want to regret not trying" is a real insight and also a convenient excuse. It can talk you into quitting a job you can't afford to quit or into a venture you haven't thought through. Use it as a tiebreaker for decisions you've already analyzed, not as a license to skip the analysis.
Survivorship bias creeps into the retelling. Bezos ran this exercise, tried, and built Amazon, so the story sounds like proof the framework works. But plenty of people minimized regret, tried, and failed. That doesn't make the framework wrong. Minimizing lifelong regret is a legitimate goal even when the attempt doesn't pay off. Just don't confuse "I won't regret trying" with "this will succeed." They're different claims.
Finally, the framework assumes your future self is a good judge. Projecting to 80 is a heuristic, not an oracle. You're guessing at what will matter to a person you haven't met yet. The guess is usually better than short-term panic, which is the whole point, but it's still a guess. Treat it as a strong signal, not a certainty, and pair it with real analysis of costs and odds. Decision models like the ones in how to apply Thinking in Bets help you keep both the emotional and the probabilistic sides honest.
How to Run Your Own Regret Minimization Exercise
Here's a practical version you can use this week on a decision you've been avoiding.
Start by writing the decision as a single sentence. Force yourself to name both paths. Writing it down does something thinking about it doesn't: it stops the problem from shape-shifting every time you turn it over.
List the short-term fears explicitly. Money, status, what people will think, the awkwardness of change. Name each one. These are the forces the framework is designed to shrink, and they lose some power once they're on paper instead of looping in your head.
Now write two short letters from your 80-year-old self, one for each path. In each, describe how that older version feels looking back. Be specific. Vague optimism doesn't count. If one letter is full of quiet peace and the other is full of "I always wondered," you have your answer.
Separate the attempt from the outcome. Ask directly: if I try this and it fails, will I regret having tried? For most genuine one-way-door decisions, the answer is no. That single question dissolves a surprising amount of fear, because fear is almost always about failure, and the framework tells you failure isn't the thing you'll regret.
Set a decision deadline. Bezos gave himself about 48 hours. Open-ended deliberation isn't wisdom, it's avoidance wearing a lab coat. Pick a date, do the exercise, and decide.
Capture your reasoning before you know the result. This last step is the one almost everyone skips, and it's the one that turns a single decision into a lifelong skill. Write down not just what you chose but why, what you feared, and what you predicted. You're leaving evidence for your future self to grade.
While you research the decision itself, keep your inputs in one place. If you're weighing a career move or a business idea, you'll read essays, watch talks, and skim case studies. Glasp's web highlighter lets you highlight the passages that actually move your thinking as you go, and the same habit applied to video means you can summarize the YouTube talks you're learning from instead of rewatching them. The goal is to decide from a rich, saved base of thinking rather than from a blank, anxious mind.
Turning One Decision Into a Reflection System
A single regret exercise helps you make one choice. A system for capturing your decisions helps you get better at making all of them.
The problem with big decisions is that you rarely get feedback on your own judgment. You choose, years pass, life moves on, and you never go back to check whether your reasoning was sound or just lucky. Without that loop, you can make important decisions for 40 years and never actually improve at it.
The fix is a decision journal. Every time you face a real one-way door, write an entry: the decision, the date, your regret analysis, what you feared, and what you predicted would happen. Months or years later, reopen it. Were your fears real? Did the thing you dreaded even come to pass? You'll almost always find that the short-term terrors were smaller than they felt, which is exactly what Gilovich and Medvec's research predicts, and that knowledge makes the next hard decision easier.
Your reading feeds this loop. The mental models you collect, from Bezos on regret to others on probability and reversibility, become the vocabulary you reason with. Saving and organizing what you read is the same muscle as keeping a commonplace book, an idea we cover in the digital commonplace book. The highlights you keep build into a personal library of thinking tools, and because you can ask an AI chat questions across your own highlights, you can pull up "everything I've saved about decision making" the moment a hard choice lands on your desk.
There's a compounding effect here that mirrors how great founders operate. They don't just make good calls once. They build systems that make their future calls better. The regret framework gives you the model. A reflection habit gives you the reps. Together they turn decision making from a nerve-wracking event into a skill you can actually sharpen. For a deeper look at that mindset, how to do great work is a natural companion read.
Frequently Asked Questions
What is Jeff Bezos's regret minimization framework?
It's a decision-making tool: imagine yourself at age 80 looking back on your life, then choose the option that leaves you with the fewest regrets. Bezos described it in a 2001 Academy of Achievement interview and used it in 1994 to decide whether to leave his job and start Amazon. The core insight is that over a lifetime, people regret the things they didn't try far more than the things they tried and failed at.
Is the regret minimization framework backed by research?
The underlying psychology is. Thomas Gilovich and Victoria Medvec found in the 1990s that people regret their actions in the short term but regret their inactions much more over the long run. By projecting to age 80, the framework deliberately moves a decision into that long-term frame, where the regret of never trying weighs heaviest. Bronnie Ware's work with dying patients points the same way.
When should I not use the regret minimization framework?
Skip it for small, reversible, everyday decisions. Your future self won't care which phone you bought or which restaurant you picked. Bezos's own Type 1 versus Type 2 framing is the filter: use regret minimization only on Type 1 decisions, the consequential one-way doors you can't easily undo. For reversible choices, decide fast and adjust.
Doesn't this framework just encourage risky decisions?
It can be misused that way, but that's not how Bezos applied it. He'd already spotted a real trend, analyzed products, and picked books before he ran the regret test. The framework is a tiebreaker for decisions you've thought through, not a license to skip the thinking. It minimizes regret over a lifetime, which is a legitimate goal even when a given attempt fails.
How is regret minimization different from expected value?
Expected value weighs outcomes by their probabilities, which works well for repeatable bets with knowable odds. Regret minimization is built for rare, one-shot, identity-shaping decisions where the odds are mostly guesses and the emotional stakes are high. They answer different questions, and thoughtful decision makers keep both in their kit.
Conclusion: The Decision You Won't Regret
The regret minimization framework endures because it does one thing exceptionally well. It quiets the loud, short-term voice of fear and lets the quiet, long-term voice of your future self be heard. That voice, the research shows, cares far less about failed attempts than about attempts never made.
You don't need to be starting a company to use it. The next time you face a real one-way door, a job, a move, a relationship, a project you keep postponing, run the exercise. Write the decision down. Ask what your 80-year-old self will regret. Separate trying from succeeding. Then decide, and save your reasoning so you can learn from it later.
Start building the base of thinking those decisions draw on. Highlight the essays and books that sharpen your judgment with Glasp's web highlighter, summarize the talks you learn from, and keep it all in one place you can question later. The best decision, at 80, is almost never the safe one you can explain. It's the one you had the courage to try.