Pre-Commitment and Post-Hoc Rationalisation: Why You Must Set Criteria Before You See Options

You’re choosing a piece of capital equipment that has to last the next decade. Call it a production line, a fleet, a core software platform, the kind of purchase that comes with a six-figure price tag and a contract you’ll be living inside long after the salesperson has moved on. You line up three vendors. The second one gives you a demo that genuinely lands. The interface is clean, the rep answers every question without flinching, and the reference customer raves on the call. By the time you sit down to score them, you find yourself leaning hard on “ease of use” and “vendor responsiveness.”

Here’s what you don’t notice. A month ago, before any demo, you’d have told anyone who asked that throughput and total cost of ownership over ten years were what mattered. Those were the numbers that justified the spend. Now they’ve quietly slid down your list, and the soft, impressive stuff has climbed to the top. The evaluation rewrote your criteria while you thought you were applying them, and it felt like clarity the whole way.

The evidence

The reason this is so hard to catch is that the distortion happens while you’re deciding, not after. Track people as information about each option arrives one piece at a time, and a consistent pattern shows up. The moment a slight preference forms, often after just the first or second fact, every later piece of information gets bent toward whatever’s already leading. People don’t only weigh the new data in favour of their front-runner. They read ambiguous data as supporting it. This is predecisional distortion, and the name matters, because it tells you when the damage is done. It happens all the way through, with each new fact you take in, long before you reach the moment you think of as the decision.

That timing closes off the obvious defence. You can’t sleep on it or think it through more carefully, because the thinking itself is the channel the distortion runs through. The longer you evaluate with a live favourite, the more lopsided the picture gets, until the case for the front-runner feels overwhelming. It feels that way because you built it, fact by fact, without ever deciding to.

A second finding makes this worse. People have strikingly little access to what actually drove their own choices. Ask someone why they picked one option over another and you’ll get a confident, reasonable-sounding answer that often has little to do with what really moved them. The answer is sincere. They’re simply constructing a tidy story after the fact for a process they couldn’t watch from the inside. So when you explain to your boss why this vendor is the right call, the explanation is sincere and may still be a rationalisation assembled to fit a preference you formed in the demo room.

There’s also a finding about what you can even see in an option. Attributes that are easy to judge on their own, a familiar brand name, a headline price, a glossy demo, carry far more weight when you look at each option separately. Attributes that only mean something in comparison, like real throughput under your actual load or the true ten-year cost once you fold in maintenance and downtime, tend to get ignored unless you’ve named them up front and you’re scoring every vendor against them side by side. This is the evaluability problem, and for a capital purchase it’s expensive. The numbers that should dominate a ten-year decision are exactly the ones that go quiet when you judge each pitch on its own merits.

How it works

Underneath all of this is a simple pull. Holding three serious options in real tension is uncomfortable and tiring. Your mind resolves the discomfort by promoting one of them early, then gathering reasons that justify the promotion. This is bolstering. Once a front-runner exists, you start attending selectively: you chase the facts that flatter it, wave away the ones that don’t, and inflate the importance of the things it happens to be good at. By the time you formally “decide,” the real deciding is behind you, and the meeting where you weigh it all up is mostly ceremony.

So the leverage sits in the sequence rather than in evaluating harder. If you fix what success looks like before a single vendor has shown you anything, you create a reference point the demos can’t move. The impressive interface still impresses you, but now it has to earn its place against throughput and cost that you committed to in writing while you were still cold. You’ve arranged the decision so the demo never gets to set the terms, which is easier than trying to resist it once it has.

By the time you sit down to weigh the options, the weighing is already bent, because you bent it yourself one fact at a time, and the only stretch where your judgement is still straight is before the first fact arrives.

How to use it

Before you take a single demo or open a single proposal, write down the three things that would make this purchase a success in ten years, and rank them. For the equipment buy that might be throughput at your real volumes, total cost of ownership including service and downtime, and integration with what you already run. Ranking is the part people skip and the part that does the work. “Throughput beats cost, cost beats integration” gives you a rule that resolves a tie, and ties are where these decisions stall and then default to whoever demoed best.

Keep it to three, maybe four. Stretch to seven or eight criteria and everything reads as important, which means nothing is, and you’re back to deciding on feel while a spreadsheet provides cover. The point of the short ranked list is to be specific enough, and fixed enough, that you notice when you start drifting off it, which a list long enough to capture every nuance never lets you do.

Write it down literally, on paper or in a doc you can’t quietly edit. An unwritten criterion is soft. It bends to fit whichever vendor you liked walking out of the room, and you won’t feel it bend. A written one is a fixed marker. You can still revise it, but now the revision is a decision you’re making on purpose rather than a drift you never noticed. That difference, between a criterion you can see slipping and one that slips in the dark, is most of the value.

Then run a check the day after you choose, before you sign. Put the criteria you actually leaned on next to the ones you wrote a month ago. If “vendor responsiveness” climbed into your top three when it started fifth, sit with that for a minute. It doesn’t automatically mean you chose wrong. Maybe responsiveness genuinely matters more than you first thought, and your original list was thin. Or maybe the demo did its job on you and you’re about to spend six figures on a feeling. Knowing which of those is true is worth the ten minutes, and it sharpens how you frame the next big buy.

Why it matters

Most procurement runs options-first. A shortlist appears, the vendors present, and the criteria take shape during the discussion, shaped by who presented most confidently and whatever stood out that week. It feels efficient, because you’re dealing in concrete options instead of abstract requirements. It’s also the exact structure that lets predecisional distortion do its work, and on a long-lived capital asset the cost of getting it wrong compounds for years.

The alternative asks for a few minutes of awkward abstraction up front. You have to answer “what would make this a good decision in a decade?” before you know what’s on offer, and before you know, the question feels unproductive. It’s actually the one stretch of the process where your judgement is still clean, before any vendor has had the chance to tilt it. Almost every purchase you’ve looked back on and thought “I weighted the wrong things” was decided in an order that let the options write the criteria. The repair is to decide what matters while you still can, before the first demo gets a vote.