← Archives02 · 17th JULY 2026
Interaction Design

The best interface is the one they've already used a thousand times

Jakob Nielsen's observation is almost embarrassingly simple: users spend most of their time on products that aren't yours, so they arrive already expecting yours to work the same way. Every deviation from that expectation is a small tax, charged before your actual idea has been judged.

Written by HET RAJYAGURU
The best interface is the one they've already used a thousand times
Every unfamiliar pattern is a small tax on someone's attention, charged the moment they land on your screen — before they've even decided whether your idea is any good.
(01)

Nobody arrives at your product with a blank slate

By the time someone opens your app for the first time, they've already learned, from dozens of other products, that the cart icon lives top-right, that blue underlined text is clickable, that a magnifying glass means search. None of that muscle memory came from you — you're borrowing it, whether you meant to or not.

Every place your interface breaks that muscle memory costs the user a small amount of conscious effort to re-learn, spent before they've formed any opinion about whether your product itself is any good.

(02)

Where "innovative" quietly turns into "confusing"

There's a real tension here: a product that looks exactly like every other product is forgettable, but a product that breaks convention everywhere is exhausting to learn. Neither extreme is the goal.

The resolution isn't "never innovate." It's knowing which parts of your interface are actually your idea, and which parts are just plumbing. Nobody chose your product for its dropdown menu. Spend your novelty budget on the one interaction that's genuinely new, and borrow convention for everything around it.

(03)

A quick test for telling the two apart

Ask, for any given pattern: if I changed this, would someone notice they were learning something new, or would they just feel lost? Navigation, form layout, standard icons — changing these rarely reads as "innovative." It just reads as friction.

The actual differentiator — the one interaction that makes your product worth choosing — is the one place where breaking convention is worth the cost of re-teaching it.

In practice
01

Default to convention for anything that isn't your product's actual idea — navigation, forms, standard icons.

02

Reserve your "this works differently" budget for the one interaction that's genuinely new.

03

If someone has to relearn something basic to use your product, that's goodwill spent, not creativity earned.

Familiarity isn't a lack of imagination. It's the thing that buys your actual idea enough attention to be judged on its own merits.

More from the archive

View all →