Most decisions are posed as yes or no when they were never that shape

The set of options is an assumption

A decision arrives looking like a fork. Take the job or stay. Buy the machine or keep repairing the old one. Confront the person or let it go. The fork feels like a feature of the situation, but it’s nearly always an artefact of how the question was first phrased — often by somebody else, often in passing, often before anybody had thought about it carefully.

Descriptive work on how organisations and individuals actually decide has repeatedly found that a large share of decisions are considered in this whether-or-not form, with a single course of action evaluated against doing nothing. Decisions considered with more than one genuine alternative on the table tend to be rated better afterwards, which is suggestive rather than conclusive — the comparison is observational and the same care that generates options may improve outcomes by other routes.

Why the set narrows so early

Generating options is effortful and evaluating them is satisfying, so attention drifts to evaluation as soon as anything is available to evaluate. Once two candidates exist, the mind switches into comparison mode and the question of what else might exist stops being asked. This happens quickly and without any feeling of having closed anything off.

The first option also anchors the frame. If someone proposes buying a replacement, the subsequent conversation is about whether to buy a replacement — its price, its timing, its model — rather than about the underlying problem, which might have several solutions that do not involve buying anything. The proposal has silently defined the category.

Ways to widen the set that actually produce options

The most reliable move is to ask what you would do if the current options were all unavailable. Forbidding the obvious answer forces a search in a different region, and the results are often not merely different but reveal what you were really trying to achieve. If both listed options were off the table, what problem would you still need to solve?

A second move is to look for the both-and version. Many apparent forks are competing for a resource that could be split, sequenced or shared — a trial of one thing before committing to the other, a smaller version of both, a reversible step that keeps the fork open for another quarter. This isn’t compromise; it is noticing that the exclusivity was assumed rather than given.

A third is to ask who else has faced this and what they did, which supplies options from outside your own imagination. The set you generate alone is bounded by what you have seen.

The difference between more options and better ones

Widening the set is not the same as maximising it, and this is where the advice can go wrong. The value comes from adding options that are genuinely different in kind, not from adding variants of the same option. Three versions of the same purchase are one option with a price slider; a purchase, a repair and a rearrangement that removes the need are three options.

This also keeps the exercise cheap. Two or three real alternatives are enough to break the yes-or-no frame, and the cost of considering them is small compared with the cost of committing to the only idea anybody had. Beyond a handful the comparison work starts to dominate, and the returns fall away.

Testing whether the fork is real

Some decisions genuinely are binary, and it is worth being able to tell. The test is whether the two options are mutually exclusive in fact rather than in framing, and whether some third party could name a third path without changing the goal. If you cannot articulate the goal separately from the options, the frame is still in charge.

The other diagnostic is time. Genuine forks usually come with a real expiry — an offer that lapses, a slot that closes — while manufactured ones come with an urgency that nobody can trace to anything specific. Asking what actually happens if this is decided in three weeks separates the two surprisingly often.

What this costs and what it buys

Generating alternatives takes time, and it introduces the possibility of dithering, which is a real cost rather than a rhetorical concession. The discipline that prevents it’s to bound the search explicitly: half an hour, three alternatives, then decide among what you have.

What it buys is that the decision is at least being made across the right field. Choosing well between two poor options is a much less valuable skill than noticing that a third existed, and the second skill is available to anyone willing to spend a few minutes before the comparison starts.