"Intuitive" is the most requested and least specified word in product design. Clients ask for it, teams promise it, and very few people can say what would make an interface intuitive rather than merely tidy.
The useful definition is narrow, and a little deflating. An interface is intuitive when it works the way the person using it already expected. That is all of it.
Which means the quality is not in your design. It is in the gap between your design and what one particular person walked in believing.
Nothing is intuitive on its own. It is only ever intuitive to someone.
Those expectations come from somewhere, and the somewhere is not your product. Jakob Nielsen named this in 2000, and it has aged better than almost anything else written about the web that year.
Users spend most of their time on other sites. This means that users prefer your site to work the same way as all the other sites they already know.
— Jakob Nielsen, Jakob's Law (2000)
Treat that as arithmetic, not advice, and it hits harder. Whatever slice of someone's screen time you get, everything else is spent being trained by other software.
Where the close button sits. What a swipe left does. That the logo takes you home, and that a form turning red means it rejected you. Decades of convention walk into your interface before you do.
Work with that inheritance and your design feels effortless. Fight it and your design feels broken, even when it is objectively better.
Why familiar usually beats clever
Anything new has to earn back the cost of teaching itself. Sometimes that is worth paying. Pinch-to-zoom was strange once, and worth learning, because nothing else came close.
But most novelty in interfaces is not solving a problem that big. It is solving the designer's problem of having seen the normal answer too many times.
Users do not share that fatigue. They see your product for a few minutes a month and have no interest in your originality.
Rule of thumb
A gear icon is not a failure of imagination
It is a sentence in a language your user already speaks fluently. Spend your novelty budget there and you have none left for the places where your product genuinely does something new.
How you know it is not working
The signal is hesitation, not failure. People rarely fail outright. They pause before clicking. The cursor circles. A tap gets undone.
Watch three people move through a flow and you will see the same half-second stall in the same spot. That stall is your interface asking someone to translate. Removing the translation is the whole job.
The tells that arrive after launch
- Support questions that repeat, in the same words.
- Features people cannot find and ask for by name — features you already built.
- Workarounds. A user who exports to a spreadsheet to do something your product already does is telling you exactly where their model and yours diverged.
- Settings changed once and never touched again, which usually means the default was wrong.
Intuitive interfaces are not the ones with the fewest elements or the most restraint, though they are often mistaken for both. They are the ones that made an accurate guess about what the user already believed, and then declined to argue with it. If you want a test for that: take any screen in your product and ask which other piece of software your user most likely learned this pattern from. If nobody on the team can name one, you are not being original — you are asking a stranger to learn something, on a Tuesday, while they are trying to do something else.