How to Make a Decision When You Are Stuck

Cut the options you would not actually accept

Start by subtracting, not selecting. Take each option in turn and ask one question: if this were chosen for me right now, with no further discussion, would I be fine with it? Anything that gets a clear no comes off the list. Do not argue with yourself about why. The reaction is the data.

Lists shrink fast under this test. Six restaurants become two. Four holiday destinations become one plus a maybe. If elimination leaves a single option standing, you were never stuck at all; you simply had not said out loud what you had already ruled out.

Name the constraint you are actually optimising for

A question like "which laptop is best" has no answer. "Which laptop is best under 900 and light enough to carry every day" has one or two. Choices only resolve against something you are trying to protect or maximise, so write that thing down in a single sentence before you compare anything.

Naming it often ends the deadlock on the spot, because one option visibly fails it. If you cannot name one constraint, you probably have two competing ones and have not decided which wins. Say which one is the tiebreaker now, while you are calm, rather than relitigating it every time you look at the list.

Compare only the differences that matter

Once you are down to two or three options, most of their attributes are identical. Both flats have a kitchen. Both jobs pay roughly the same. Shared attributes carry no decision weight, however much you enjoy reading about them, so strike them out and look only at what actually differs.

What remains is usually a short list. Work through it like this:

• Write only the attributes where the options genuinely differ.
• For each one, say which option wins, and whether it wins by a lot, a little, or barely.
• Delete any difference you would not pay anything to get. If it is not worth money, time, or effort, it is not a reason.
• If every remaining difference lands on "barely", you have reached a tie and further comparison will not break it.

Set a deadline that matches the stakes

Deliberation costs the same whatever you are deliberating about. Twenty minutes spent choosing lunch is the same twenty minutes as twenty minutes spent choosing a laptop, but only one of those is worth it. Give each decision a time budget before you start: a minute for lunch, an evening for the laptop, a week for the job offer.

This is the practical form of the idea usually called decision fatigue. Your supply of careful attention in a day is limited, and small choices drain it much as large ones do. When the deadline arrives, you decide with what you have. A decision that is cheap to get wrong should also be cheap to make.

Learn to recognise a genuine tie

Real equivalence exists and it is worth being able to spot. You have probably reached it when you have switched preference more than once, when new information stops moving you, and when you can honestly say you would be content living with either outcome a month from now.

A tie is good news, not failure. It means the analysis is complete and the gap between the options is now smaller than the cost of continuing to think about them. At that point the expensive mistake is not picking wrongly; it is not picking at all.

Be honest about what is actually stopping you

Most moments of "I cannot decide" are really "I do not want to lose the other option". That is not indecision. It is the regret of the road not taken, arriving early and pretending to be analysis. Every choice costs you the alternatives, and no amount of thinking makes that cost go away.

Test it. Decide privately that you are taking option A, sit with it for a minute, and watch the first reaction. Relief with a twinge about B is normal and means take A. Genuine sinking means your preference is B and you now know it. Sometimes the honest answer is neither, or not yet, and that is a legitimate choice too, as long as you say it deliberately instead of letting the delay make it for you.

When randomness is the right last step

If you have eliminated the unacceptable, named the constraint, compared only the differences, hit your deadline, and the options are still level, then flip something. A coin flip or a spin wheel does one useful job here: it turns a tie into an outcome without spending any more of your attention. It is the end of the process, not a way around it.

It also gives you one last piece of information. Watch how you feel when the result appears. Disappointment means you had a preference after all, so take the other one; nothing obliges you to obey a coin. Use a best-of-three or best-of-five run if a single flip feels too abrupt, or a wheel when there are more than two survivors on your list.

The condition for all of this is that the tie is real. Randomness is the wrong tool for medical, legal, financial, safety and employment decisions, or anything else where one outcome is meaningfully worse than another, because chance is only fair when the options genuinely are.