Decision-making and strategy
Regret minimization
For a big personal choice, imagine yourself at 80 looking back, and pick the option you would be least likely to regret.
Jeff Bezos, describing his 1994 decision to leave the investment firm D. E. Shaw, in a 4 May 2001 interview with the Academy of Achievement, where he called it a "regret minimization framework". It is a personal rule of thumb and is unrelated to the formal minimax regret rule of decision theory.
Bezos's version is a thought experiment. Project yourself forward to age 80, look back on your life, and ask which choice you would regret. His goal, in his words, was to have "minimized the number of regrets". From that distance, short-term costs such as a bonus given up halfway through the year look small, and never having tried looks large.
It works by changing the time frame. Near-term losses are vivid and certain, while the long-term cost of not acting is vague, so people overweight the first. Looking back from old age shrinks the near-term costs to their real size and makes the path not taken concrete. It also separates two kinds of failure: trying and failing, which Bezos said he would not regret, and never trying, which he thought he would.
The frame leans toward action, and it suits people for whom failure is survivable. Bezos had skills and savings and could have found another job; for someone whose downside is ruin, the view from 80 understates the cost. Your guess about what an older you will regret is also a forecast, and it can be wrong. Use it for rare, large, personal choices, and be wary when it is wheeled out to justify a decision already made.
Example
In 1994 Jeff Bezos was a 30-year-old vice president at D. E. Shaw in New York when he noticed that internet usage was growing by 2,300 percent a year. He told his boss he was going to "start this company selling books online". They took a two-hour walk in Central Park, and his boss said it sounded like a good idea for someone who did not already have a good job, and asked him to think about it for 48 hours. Leaving mid-year meant walking away from his annual bonus. He left, drove to Seattle, and Amazon.com opened to the public on 16 July 1995.
Where it applies
Use it for career moves: leaving a stable job to build something, joining an early-stage AI company, taking a year to study. For engineering decisions it adds little; those usually have data, and reversibility matters more than regret. A useful pairing is to ask what you would regret at 80 and then what you can afford to lose this year.
Think about it
Looking back from age 80, which choice in front of you now would you most regret not having tried?
Applied it to something real? Log it.
Test yourselfHow does imagining yourself at 80 change the way you weigh a risky choice?
It shrinks vivid near-term costs, like a lost bonus, to their real size and makes the long-term cost of never trying concrete, so you compare the regret of failing against the regret of not acting.
Related models
- One-way and two-way doorsMake reversible decisions quickly and save slow, careful deliberation for the ones you cannot undo.
- Asymmetric betsTake bets where the most you can lose is small and known and the most you can win is large, even if you expect to lose most of them.
- Pre-mortemBefore you start, imagine the project has already failed and ask everyone to explain why.
- The sunk cost fallacyMoney, time or effort already spent cannot be recovered, so it should not count in what you do next; people count it anyway.
- Expected valueWeigh each outcome by its probability and add them up; that is what a repeated bet is worth on average, and only on average.