I want to give you a quick test.
Think of a product decision you made recently — or a recommendation you gave, or a view you hold about what makes a product good. Now ask yourself: how did I arrive at that?
If the honest answer is "I've seen this kind of problem before and I remember how it resolved" — that's pattern matching. If the honest answer is "I worked through the underlying logic from scratch" — that's first principles.
Neither is bad. Pattern matching is fast and often right. First principles is slow and often overkill. The problem isn't using one or the other — it's not knowing which one you're doing.
The tell
Here's how you can tell the difference, reliably.
Pattern matching produces confident answers quickly. You recognize the shape of the problem, your brain retrieves the associated resolution, and you feel certain. The certainty is the tell. If you feel confident before you've really examined the problem, you're almost certainly pattern-matching.
First principles produces uncertainty first, then (sometimes) confidence. When you're working from the ground up, you notice what you don't know. You encounter branching points where the answer depends on a fact you haven't established. The process is uncomfortable. If your thinking isn't producing any uncertainty, you should be suspicious.
The second tell is how you respond to being challenged. Pattern-matched conclusions are often defended by appeal to precedent: "this is how it worked at Company X," "this is what the research says," "every PM I've talked to agrees." First-principles conclusions can be defended by the reasoning itself: "here's the logic, here's where it depends on what's true, here's what would change my mind."
Can you defend your position by walking through the reasoning? Or only by pointing at authorities who agree with you?
Why this matters in product work
Most product decisions are made under time pressure with incomplete information. That's the environment. Pattern matching is structurally advantaged in that environment: it's fast, it feels confident, it's easy to sell to stakeholders.
But it has a failure mode that's particularly dangerous in product: patterns learned in context A applied to context B, where the underlying conditions are different. This is how you end up with a B2B SaaS company applying the growth playbook from a consumer social product. Or a team copying Spotify's squad model without understanding the organizational prerequisites. Or — the most common version — a product team optimizing the metric that worked for the last product, applied to a completely different user population.
The pattern looks right. The context is different. The metric moves in the expected direction. The product fails the people who needed it.
A practice
I try to flag, explicitly, when I'm pattern-matching versus working from first principles. Not as self-criticism — as calibration.
When I catch myself saying "this is like that problem I saw at MakeMyTrip," I try to immediately follow it with: what makes this the same? What makes it different? If the difference matters, which framework am I actually using?
This takes maybe thirty seconds. It doesn't always change the conclusion. But it changes my relationship with the conclusion — from "I know this" to "I believe this, provisionally, for these reasons."
That shift — from knowledge to provisional belief — is where real thinking starts.
The goal isn't to never pattern-match. It's to know when you're doing it, and to have chosen it deliberately, rather than being dragged along by the comfort of recognition.