Mentavo
Two apartments, two viewings, a week to think it over. At the end comes a sentence that sounds like deliberation: "The layout is just better."
What's interesting is that the layout featured in no consideration before the viewings. The things that mattered were: commute, rent, noise.
This isn't unusual and it isn't a character flaw. Look at options first, and you'll afterwards weight whatever the more appealing option happens to do well.
One unglamorous change works against it: the order.
Why criteria invented afterwards are unreliable
Comparing two possibilities produces a leaning early – often within minutes, sometimes for reasons you couldn't name.
What follows looks like weighing up from the inside but works more like justification: features where the preferred option does well suddenly feel more important. Features where it falls short feel dispensable.
The result is a decision that feels well reasoned, and whose reasoning doesn't date from before you looked.
None of this matters much for small things. For decisions you carry for years, it's expensive.
Reversing the order
The method has four steps and rarely takes more than ten minutes.
- State the question. One sentence that names the decision exactly – not "change jobs?" but "do I stay in this role until at least March, or start applying now?"
- Write down the thresholds. Conditions without which an option is out. Three at most, or everything gets excluded.
- Choose your deciding criteria. Two or three points that actually differentiate between the remaining possibilities – plus which one outweighs the others in a tie.
- Only now look at the options. And along the written criteria, not along the impression.
Step three is the hardest, because it asks for a ranking before you know which option that ranking favors. That's exactly where its value sits.
Threshold or deciding criterion?
These two get mixed up regularly, with noticeable consequences.
A threshold is a line: an option either meets it or doesn't. "No more than 45 minutes of commuting." It isn't weighed, it's checked.
A deciding criterion is a scale: more of it is better. "How much independent work there is."
Treat a line as a scale and you end up with sentences like "a bit far, but on the other hand…" – which dissolves the very condition you'd called non-negotiable.
The reverse, treating a scale as a line, leads to premature exclusion. Which is why one sentence per criterion is worth it: is this a condition or a preference?
Why three criteria are enough
The urge to account for as many aspects as possible feels thorough. In practice it means almost every option scores well somewhere – and the differences disappear.
With two or three clear deciding criteria the comparison stays sharp. Everything that falls outside them isn't unimportant; it's just not decisive.
There's a side effect on energy: long criteria lists increase the number of comparisons, and with them the cost of every repetition. How that plays out over a day is covered in the piece on the psychology of decision fatigue.
A worked example
Question: "Do I take the project lead role or stay where I am?"
Thresholds: no permanent weekend work; a handover from somebody who knows the clients.
Deciding criteria, ranked: (1) how much time is left for hands-on work, (2) how predictable the weeks are, (3) how much I learn in twelve months.
Looking at the options, the lead role is clearly ahead on (3) and clearly behind on (1). Because (1) was ranked higher beforehand, the question is answered – which doesn't mean the answer is pleasant.
What matters is how you handle the discomfort that follows: it indicates that the rejected option contained something that counts. It isn't evidence that the ranking was wrong.
The decision note
A second, frequently skipped part: record what you expect.
Three lines will do – the criteria, the option you chose, and what should be true in three months if the decision holds.
The payoff comes later. Without a note, hindsight serves you a tidied-up version of your own reasoning; with one, you can read what actually counted. In Mentavo, a note like that stays attached to the period it came from and can later be viewed next to how things actually went.
Across several decisions, this builds something more useful than any single deliberation: an overview of which criteria have served you well – and which keep appearing without ever having been decisive.
One question before it's final
Before committing, a last check known in decision research as looking back from an imagined failure: suppose this goes wrong within a year – what was the reason?
That phrasing surfaces objections which "what could go wrong?" doesn't, because it treats the failure as given and asks only for the cause.
Often the reason named is already known – and can be defused with a small precaution that wouldn't have been taken without the question.
When the method doesn't fit
Not every decision needs criteria.
- For small choices, the structure costs more than the mistake. Choosing between two lunches is better done fast.
- For matters of taste there are no sensible criteria. Whether you like a picture can't be weighted.
- Under time pressure, a good decision in two minutes usually beats a very good one in two days.
The effort pays where the consequences last and a correction is expensive.
When both options score the same
Occasionally you hit the case the method is meant to prevent: after the comparison, two possibilities are level.
That isn't a sign of bad criteria. It usually means the decision genuinely is a tie on what counts – and then three additional questions help:
- Which is easier to reverse? In a tie, the reversible option is the more sensible one.
- Which delivers information faster? What you'll know in six weeks is worth more than what becomes clear in two years.
- Who carries the consequences if it goes wrong? If others are affected more than you are, their view belongs in the decision.
And if none of that moves the needle: in a genuine tie, timing matters more than the choice. Further deliberation then costs more than the difference between the options ever will.
What this article doesn't claim
Criteria don't make decisions right. They make them traceable – and they stop the yardstick from being adjusted to the leaning after the fact.
Nor does the piece claim intuition is unreliable. In areas where you have a lot of experience it's often accurate; there's a separate article on the science of intuition.
Conclusion: the yardstick belongs before the choice
Most hard decisions don't get easier through longer comparison. They get easier when it's settled in advance what the comparison is against.
That's uncomfortable, because you have to commit before knowing which option benefits.
Which is precisely why it works.
