If decisions you haven't made are piling up and people are waiting on you, don't try to think harder about each one. Sort them by one question: how hard would this be to undo? Decide the ones you can undo now, with what you know, and give the ones you can't undo a date and a short check before you make them. By the end you'll have the door test that sorts every open decision, a time rule for each kind, and a regret check that keeps fast from turning into reckless.
Clear the pile in seven steps: fast on what you can undo, slow on what you can't
These seven steps take one pass through every decision you're still sitting on. Steps 1 to 4 sort the pile and set the pace; steps 5 to 7 are for the decisions you can't take back. Each step says what to do, how you know it worked, and why.
Step 1: Get every open decision out of your head and onto one list
Do this. Write every decision you're still sitting on onto one page: the decision, who is waiting on it, and roughly when it first came up.
It worked when every open decision you can think of is on that one page, including the ones other people are waiting on, and none lives only in your head.
Why. In studies published in 2011, Masicampo and Baumeister found that unfinished goals caused intrusive thoughts during an unrelated reading task and poorer performance on an unrelated anagram task 1. In the same studies, letting participants make specific plans for their unfulfilled goals "eliminated the various activation and interference effects" 1. Masicampo and Baumeister concluded that committing to a specific plan for a goal "may also free cognitive resources for other pursuits" 1. We read that as the reason the list comes first: an open decision written down with a plan stops competing for your attention. In our view, decisions the people around you would rather not make can sit with them instead of reaching you, so ask your team which calls they are waiting on and add those too. Cross out any line you wouldn't miss if it stayed undecided; dropping a maybe on purpose is a decision too.
Example. One owner described coming back from four days off: "Came back to a list of things that had stalled, decisions nobody had made, and two clients who had been waiting on me specifically to move forward." The same owner added: "Nothing catastrophic, but nothing had moved without me either!" We read that as the cost of decisions only you hold: they waited on one person, and nothing moved.
Step 2: Mark each decision undo or no-undo
Do this. Next to each line write one word: undo if you can say in one sentence how you would reverse it at a cost you can carry, no-undo if you can't.
It worked when every line carries one of the two words, and for each undo line you could say out loud how you would reverse it.
Why. In his 2015 letter to shareholders, Jeff Bezos described some decisions as "consequential and irreversible or nearly irreversible" and called them one-way doors 2. Bezos wrote that one-way-door decisions "must be made methodically, carefully, slowly, with great deliberation and consultation" 2. Bezos wrote that most decisions are "changeable, reversible", which he called two-way doors, and that with a poor one "you don't have to live with the consequences for that long" 2. Bezos wrote that two-way-door decisions "can and should be made quickly by high judgment individuals or small groups" 2. In our view, these two labels are all you need for the whole pile, because the label decides how much time each line gets.
Example. One person weighing whether to pay a referral source 15% put it this way: "Part of me thinks a warm pipeline costs money and this is just sales without the salary. Part of me thinks once I open that door I am renting my book forever." We read a choice like that as a door question in disguise: if the first payment could be stopped later, it is an undo decision, and the debate can end with a small trial.
| Undo (two-way door) | No-undo (one-way door) | |
|---|---|---|
| The test | You can say in one sentence how you would reverse it | You can't, or reversing it would cost more than you can carry |
| Information | What you have now | The one fact you still need |
| When | Within a day | On a set date |
| Before you decide | Nothing more | A premortem and a regret check |
Step 3: Decide every undo item within 24 hours with what you know now
Do this. Decide every undo item within 24 hours of making the list, using the information you have today, and write the choice next to the line.
It worked when within 24 hours every undo line has a written choice next to it and none is marked as waiting for more information.
Why. In his 2016 letter to shareholders, Bezos wrote that "most decisions should probably be made with somewhere around 70% of the information you wish you had" 3. Bezos wrote that if you wait for 90% of the information, "in most cases, you're probably being slow" 3. Bezos wrote: "If you're good at course correcting, being wrong may be less costly than you think, whereas being slow is going to be expensive for sure" 3. Our rule of thumb is that an undo decision needs less certainty than that, because if you can reverse it, being wrong costs one correction. Make small calls that don't need a line on the spot, and if the pile feels too big to start, clear the quickest undo items first to get moving.
Step 4: Give every no-undo item a date and the one fact you still need
Do this. Give every no-undo line a decision date within 14 days and write next to it the one fact you still need before that date.
It worked when every no-undo line has a date no more than 14 days out and one named fact, and you know who will bring you that fact.
Why. In our view, more time does not improve a decision past a point; a fact does, so the line names the fact instead of asking for time. We read the date and the named fact as the start of a specific plan for that line. If the fact can't arrive before the date, decide on the date without it.
Example. One person whose father gave them his business a year ago wrote: "Now my mother is saying I should at least run it for another 5 years before making a decision." In our view, five years is a horizon, not a decision date; a no-undo choice like that can still get a date in the next two weeks for one smaller step, such as which fact to collect first.
Step 5: Before a no-undo date, assume the decision has already failed
Do this. A day or two before each no-undo date, imagine it is a year later and the choice has failed, write the three likeliest reasons, then write the worst case and what you would do next.
It worked when each no-undo line has a short written list of failure reasons and a worst case you've decided you could live with, or the choice has changed because of them.
Why. In a premortem, as Gary Klein described it in Harvard Business Review in 2007, "team members assume that the project they are planning has just failed" and then generate plausible reasons for that failure 4. Our advice is to use the same move on your own: you are the team, and the project is the choice on your list. In our view, walking the worst case to the end, including what you would do next, turns a vague fear into a plan you can judge. If the reasons keep circling the same two options, look for a third before the date.
Step 6: On the date, pick the choice you'd regret less, and write down why
Do this. On the decision date, ask which choice you would regret more looking back in a few years, pick the other one, and write one sentence of reasoning next to the line.
It worked when on the date the line is closed with a choice and a written reason, not moved to a new date.
Why. A 1995 review in Psychological Review by Gilovich and Medvec reported that actions, or errors of commission, "generate more regret in the short term; but inactions, or errors of omission, produce more regret in the long run" 5. We read that as a warning about the pile: the lines you leave open are the kind of inaction that review links to regret in the long run. In a large randomized field experiment reported by Steven D. Levitt in 2016, people having difficulty making a decision flipped a coin to help determine their choice 6. Levitt found that people told by the coin to make a change were much more likely to make one and were happier six months later than those told to keep the status quo 6. Levitt wrote that the results "suggest that people may be excessively cautious when facing life-changing choices" 6. In our view, that's a reason to let the date force the call instead of waiting to feel certain.
Example. One owner of 15 years said of people decisions: "my only regret has ever been not firing them sooner." Another owner said the same about a delayed decision: "Once I let that person go the only regret I had was that I didn’t do it sooner." Both were about letting someone go, which our article on managing underperforming employees covers; here they show only how a delay can look in hindsight.
Step 7: When your team is split, disagree and commit instead of waiting it out
Do this. When a decision on your list involves your team and you still disagree on the date, say where you stand, make or confirm the call, and ask each person, you included, to commit to it.
It worked when the date passes with a decision each person has said they will back, and no one reopens it before a review date you set.
Why. Bezos recommended the phrase "disagree and commit" and wrote that it "will save a lot of time" 3. Bezos wrote that the rule applies to him too: "If you're the boss, you should do this too. I disagree and commit all the time" 3. Bezos wrote that when teams are deeply misaligned and nobody escalates, the default way the dispute gets settled "is exhaustion", because "whoever has more stamina carries the decision" 3. We read the date as a way to stop that: the call is made on the date, not won by whoever lasts longest.
Speed can be safe when it matches how hard a decision is to undo
Bezos wrote that as organizations get larger, "there seems to be a tendency" to use the heavy-weight process meant for one-way doors on most decisions, and that the result is "slowness, unthoughtful risk aversion, failure to experiment sufficiently, and consequently diminished invention" 2. In a footnote to the same letter, Bezos added that companies which habitually use the light-weight process to make one-way-door decisions "go extinct before they get large" 2. We read the two warnings together: the risk is not speed or care, it's using one pace for every decision. Bezos closed the 2016 section on decisions by asking: "have you settled only for decision quality, or are you mindful of decision velocity too?" 3.
Every open decision waits in the owner's head until it feels right, and the people who need it wait too.
Every open decision has a door type and a date: the ones you can undo are made now, the ones you can't are made on their date after a check.
In our view, advice on being decisive often treats every decision alike: either trust your gut or slow down and think. Neither tells you which of your open decisions deserves which, and neither puts a date on anything. The door test and the date are the two pieces we think the gut-or-slow advice is missing.
The move this article covers is from every open decision waiting in your head until it feels right, to every open decision carrying a door type and a date. Handing decisions to other people is the next move, and it has its own article.
Bezos used a regret test to start Amazon, and it cost him his annual bonus
Jeff Bezos said he made the decision to start Amazon with what he called a "regret minimization framework": he projected himself forward to age 80 and asked which choice would leave him with the fewest regrets 7. Bezos said in a 2001 interview: "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" 7. Bezos said that leaving his Wall Street firm in the middle of the year meant walking away from his annual bonus, "the kind of thing that in the short-term can confuse you" 7. In the same interview Bezos said: "Most regrets, by the way, are acts of omission and not commission" 7. We read the story as the regret check from step 6 used on a true one-way door: a loss he could see in the short term, weighed against the regret he expected looking back at 80.
Fast decisions can still go wrong, and four worries deserve an answer
If you worry that a fast decision will turn out badly, separate the decision from the outcome. The publisher describes Annie Duke as "a former World Series of Poker champion turned business consultant" 8. The publisher's description of Annie Duke's Thinking in Bets (2018) says professional poker players accept that "great decisions don't always lead to great outcomes and bad decisions don't always lead to bad outcomes" 8. The same description sums up thinking in bets as asking: "How sure am I? What are the possible ways things could turn out? What decision has the highest odds of success?" 8. In our view, that's the answer to the worry: judge a decision by the steps it followed, not only by how it turned out, and use the written reason from step 6 to tell the two apart.
If you worry that you'll settle for less than the best option, weigh what holding out costs. A 2002 paper by Schwartz and colleagues found, across seven samples, negative correlations between the desire to maximize and happiness, and positive correlations between maximizing and regret 9. In the same paper, maximizers were less satisfied with consumer decisions than satisficers 9. We read that as a reason to stop once an option clears your bar on an undo decision, instead of hunting for the best one.
The weight of that reality makes every decision feel even more overwhelming.one owner with a failing business
In our view, the method still holds when the stakes are high; pressure is a reason to keep the list, not to drop it. For no-undo decisions with legal, tax or large money stakes, the one fact you still need may be an adviser's opinion, so put that opinion before the date.
If you've heard that you should only decide in the morning because judgment wears out, look at one study of rulings through the day. A 2011 study of sequential parole decisions by Danziger, Levav and Avnaim-Pesso reported that favorable rulings dropped from about 65% to nearly zero within each session and returned to about 65% after a food break; the finding was challenged in print the same year and the authors replied, so it is not settled 10. Weinshall-Margel and Shapard challenged that study in a published letter, reporting that case order was not random and that unrepresented prisoners usually went last in each session and were less likely to be granted parole 11. Weinshall-Margel and Shapard concluded that the pattern of favorable decisions peaking after a meal break "is likely an artifact of the order of case presentation" 11. We read that as too unsettled to plan around; our advice is to put your effort into the dates on your list, not the time of day.
Put every open decision on one list today and count what's left in 14 days
Set aside time today to build the list from step 1 and mark every line undo or no-undo. Decide every undo line within 24 hours. Give every no-undo line a date within 14 days and the one fact you still need. In 14 days, count the open lines and compare them with your first count.
your open-decisions list
You now have the door test, a time rule for each kind, and a regret check
You started with a pile of decisions waiting on you and nothing moving. The door test sorts that pile into what you can undo and what you can't. The time rule decides the undo items within 24 hours and gives every no-undo item a date within 14 days and the one fact it needs. The regret check, with a premortem before it, lets you make the no-undo calls on their date without rushing them.
Read before: How to Replace Yourself in Your Business (Before You Can Afford a Team). Read next: How to Stop Micromanaging Your Team: Delegate Decisions, Not Tasks.
If the decision you keep delaying is about a person, read How to Manage Underperforming Employees (Before You Decide to Fire Them).
See where your business stands on decisions: take the Scorecard.
Sources
- Consider it done! Plan making can eliminate the cognitive effects of unfulfilled goals · Journal of Personality and Social Psychology (Masicampo & Baumeister), October 2011
- 2015 Letter to Shareholders · Amazon.com, Inc. (Jeff Bezos), April 2016
- 2016 Letter to Shareholders · Amazon.com, Inc. (Jeff Bezos), April 2017
- Performing a Project Premortem · Harvard Business Review (Gary Klein), September 2007
- The experience of regret: what, when, and why · Psychological Review (Gilovich & Medvec), April 1995
- Heads or Tails: The Impact of a Coin Toss on Major Life Decisions and Subsequent Happiness · NBER Working Paper 22487 (Steven D. Levitt), August 2016
- Jeff Bezos interview, King of Cyber-Commerce · American Academy of Achievement, May 4, 2001
- Thinking in Bets: Making Smarter Decisions When You Don't Have All the Facts · Penguin Random House (Annie Duke, Portfolio), 2018
- Maximizing versus satisficing: happiness is a matter of choice · Journal of Personality and Social Psychology (Schwartz et al.), November 2002
- Extraneous factors in judicial decisions · PNAS (Danziger, Levav & Avnaim-Pesso), April 2011
- Overlooked factors in the analysis of parole decisions · PNAS (Weinshall-Margel & Shapard), October 2011
