The first path you teach becomes the only one they see

There’s a classic experiment where people are handed three jugs of fixed sizes and asked to measure out an exact amount of water. The first few puzzles all yield to the same slightly fiddly recipe — fill the big jug, pour off the middle one, pour off the small one twice. People learn it, and it works, and they feel clever. Then comes a puzzle that the same recipe still solves — but that also has a dead-obvious two-step shortcut sitting in plain sight. Almost everyone grinds through the long recipe and never notices the shortcut. And here’s the tell: people shown only the shortcut puzzle, with no long recipe drilled into them first, spot the easy solution immediately. The long way didn’t make them smarter. It made them blind.

Psychologists call this the Einstellung effect — a mental set, a groove worn by the first method that worked. It’s a cousin of functional fixedness, the reason a person hunting for something to prop a door will overlook the heavy stapler on the desk because a stapler is “for stapling.” Once the brain has a working answer, it quietly stops searching. This is not stupidity; it’s efficiency. A brain that re-derived the best approach from scratch every time would never get anything done, so it caches the first success and reaches for it reflexively. The cost of that cache is invisible precisely because you stop looking — you can’t miss a shortcut you never searched for.

Now put that brain in front of your product on day one. Whatever workflow a new user stumbles into first — the way they happened to accomplish the thing that made them go “oh, I get it” — is about to become their permanent model of how your product works. Not because it’s the best path. Because it’s the first one that worked. They will repeat it, teach it to teammates, and, months later, actively resist the cleaner path you eventually ship — because from inside the groove, the old way doesn’t feel clumsy, it feels like how the tool works. You didn’t just teach them a task. You installed a mental set, and mental sets are sticky.

So the build decision is quieter and more consequential than it looks: the first-run experience is not an introduction, it’s the permanent one. Decide which path you want people using in a year, and make that the path they discover first — before any workaround has a chance to set. Three practical consequences follow. First, don’t let the clumsy way be the easy way to stumble into; if there’s a good path and a hacky path, the good one has to be the one that’s obvious on day one, because whichever they hit first wins. Second, a “better way” shipped later is not competing with the old way on merits — it’s competing with a worn groove, so it needs far more than a changelog note; it needs a reason to break set, ideally the old path gently closing so there’s nothing to fall back into. Third, treat the moment of first success as precious real estate: the exact sequence a user runs to reach their first win is the sequence they’ll marry.

Two honest cautions. Set isn’t pure villainy — it’s also what makes experts fast. A seasoned user’s reflexive groove is their fluency, and a product that keeps forcing people to re-decide the basics is exhausting; the goal is to make the first-learned groove a good one, not to prevent grooves. And the fix is not to rail users so hard they can’t explore — that provokes its own backlash (people resent a tool that won’t let them find their own way). The move is subtler: make the efficient path the path of least resistance at the exact moment the mental model is forming, and spend your “break their set” energy sparingly, on the few changes worth the fight.

The science, to look up: the Einstellung effect from Abraham Luchins’ water-jar experiments (1942), showing that a drilled-in method blinds people to a simpler available one; Karl Duncker’s work on functional fixedness (the candle problem); and the broader problem-solving literature on set and fixation. The effect is robust and easy to demonstrate; treat it as a well-established tendency, not an on/off switch — some people and some framings break set more readily than others.