
You now know that a number supports a claim rather than proving it. But before you weigh how much support a figure gives, there is an earlier question that decides whether it belongs in the conversation at all: does this evidence actually bear on the decision you are making? A number can be perfectly accurate and still be beside the point, and spotting that is the first of the four questions.
Some numbers feel relevant the moment you see them. They are on-topic, they come from a trustworthy-looking source, and they point in a clear direction, so it is tempting to let them settle the matter. The trouble is that “sounds relevant” and “is relevant” are not the same thing. A figure can be about roughly the right subject while measuring something that does not actually answer the question in front of you.
Think of a shopper choosing between two blenders who reads that one brand “sold two million units last year.” That number is real, and it sounds like a point in the brand’s favor. But the decision is which blender works better for me, and units sold measures popularity, not performance. The figure is on-topic enough to feel useful, yet it does not bear on the choice being made. Catching that gap, before you give the number any weight, is what judging fit is about.
To test fit, you need to be clear about three plain things.
The single most useful move in this chapter is to ask, of any number, “what does it actually measure?” and then hold that answer against the decision you named. Most fit failures come from a quiet mismatch between the two: the number measures one thing, the decision needs another, and the resemblance between them hides the gap.
Isn’t a number that is roughly about the right topic good enough for me to use?
Not on its own. Being about the right topic is what makes a number worth checking, not what makes it fit. The test is whether what it measures lines up with what your decision actually turns on.
When a number lands in front of a decision, put it through three quick questions before you lean on it.
What exactly does this number measure? Name the precise thing it counts, not the general topic it belongs to.
What does my decision actually turn on? Name the specific thing you would need to know to choose well.
Do those two line up? If what the number measures is the thing the decision turns on, it fits. If it measures a stand-in, a symptom, or a neighbor of that thing, treat it as, at best, partial.
A number that passes all three earns a place in your appraisal. One that fails the third question can stay in the conversation only if you are honest that it speaks to something adjacent, not to the decision itself.
Picture someone deciding whether to pay for a gym membership. To settle it, they look at their fitness tracker and see they average 9,000 steps a day, comfortably active. “I am clearly an active person,” they reason, “so the membership will get used.” The step count is accurate, personal, and squarely on the topic of fitness. Does it fit the decision?
Run the check.
What does the number measure? Steps taken during ordinary daily life, walking to work, moving around the house, running errands.
What does the decision turn on? Whether they will actually go to a gym and use it enough to justify the cost.
Do they line up? Not really. A high step count shows everyday movement, which is exactly the kind of activity a person can do without ever entering a gym. It may even count against the membership: someone already active on their feet might have little reason to pay for one.
So the step count, accurate as it is, does not bear on the decision the way it first appeared to. Evidence that would fit looks different: how often they have kept up past subscriptions, whether the activities they enjoy actually need gym equipment, how far the nearest gym is. The point is not that the step count is a bad number. It is a fine measure of the wrong thing for this choice.
That is the whole move. Name the decision, name what the number measures, and check whether the two genuinely meet, before you let any figure earn its weight. Once a number clears the fit test, you can turn to what it quietly assumes.
A number can be accurate and on-topic yet still fail to bear on the decision you are making.
Judging fit means separating evidence that is relevant from evidence that only sounds relevant.
The core move is to ask what a number actually measures and hold that against what the decision truly turns on.
Fit fails when a number measures a stand-in, a symptom, or a neighbor of the thing the decision needs.
Evidence that clears the fit test earns a place in your appraisal; evidence that does not should not carry weight.
A number that fits the decision still rests on things it quietly takes for granted; the next chapter surfaces those hidden assumptions and the limitations they expose.