Herbert Simon's observation in 1956 was that the economic model of a decision — consider every option, evaluate each against a stable set of preferences, pick the best — describes something no human being has ever done, because gathering and comparing options costs time and attention that the model treats as free.

What people do instead is set a threshold and take the first thing that clears it. Simon called it satisficing, from satisfy and suffice, and his point was that this is not a failure to be rational. Under real constraints it is the rational procedure, because the search has to be paid for out of the same budget as everything else.

The two dispositions

The research that followed split people roughly into two habits.

Maximizers want the best available option and keep looking until they are reasonably sure they have it. They read every review, they check one more supplier, they keep the tab open.

Satisficers decide in advance what would be good enough and stop at the first thing that qualifies.

The consistent finding is uncomfortable for maximizers: they get slightly better outcomes by objective measures and feel worse about them. The better outcome is real — the extra search does find something marginally better — and it is swamped by the cost of getting it, and then by the awareness that something better might still be out there. Having chosen well and knowing you might not have is a worse state than having chosen adequately and having stopped.

Why the arithmetic favors it more than it looks

Three costs, and only the first is usually counted.

The search itself. Hours, and attention that was available for other things — which is opportunity cost, applied to the one resource nobody can borrow more of.

The decline in what searching finds. Option quality does not improve linearly with effort. The first few candidates capture most of the available range and the rest are refinements, which is diminishing marginal utility operating on the search rather than on the thing.

The cost of the choosing itself. A long comparison degrades the quality of the judgment at the end of it, because deliberation depletes. See decision fatigue — the last comparison in a long day is not made with the same instrument as the first.

Setting the threshold, which is the actual skill

Satisficing is only better than settling if the threshold is set deliberately, before looking, and in terms that can be checked.

Before looking matters most. A standard chosen after seeing the options is not a standard; it is a description of the best option, and it will be revised again when a better one appears. Writing it down first is what keeps the search from being endless.

Checkable matters nearly as much. "A good apartment" is not a threshold. "Under this rent, within this commute, with a room I can work in" is one, and the first place meeting all three is the answer.

When to maximize instead

The distinction is not a personality test, it is a property of the decision, and two things put a decision in the other category.

Irreversibility. If you cannot undo it, the extra search is cheap relative to being stuck. Satisfice on things you can change and take your time on things you cannot — which is the same line drawn by reversible decisions.

A wide spread of outcomes. Where the best option is dramatically better than an adequate one — a co-founder, a surgeon, where to spend a decade — the tail is worth searching for. Where the options are close together, and most are, the search is buying almost nothing.

Most decisions fall on the satisficing side and are treated as though they do not. The error is expensive precisely because it is invisible: nobody records the hours spent choosing among equivalents, and the person doing it experiences it as being careful.