The UX laws worth knowing, and when each one applies

There are more lists of UX laws than there are UX laws. Most are alphabetical, which is the least useful possible order — it puts Fitts’ Law next to the Peak-End Rule because both begin near the front of the alphabet, and separates the four principles that all describe the same thing.

These are grouped by the decision each one helps you make. If you are arguing about whether two things look related, four of them apply and the rest do not. If you are arguing about how many options to show, a different four apply. Knowing which set you are in is most of the value.

What the eye groups before you have read anything

Four Gestalt principles decide what a layout means before conscious attention arrives. They are not stylistic preferences — they describe what has already happened by the time someone starts reading.

• Gestalt psychology — the framework the rest sit inside.
• The Law of Proximity — distance groups things, whether or not they belong together.
• The Law of Similarity — shared colour or shape reads as shared function.
• Common Region — a boundary overrides distance.
• Uniform Connectedness — an explicit visual link is the strongest grouping signal of the four.
• The Law of Prägnanz — ambiguity resolves to the simplest available reading.

How much you are asking someone to hold in their head

These four are about cognitive load: the cost of deciding, remembering and interpreting. They are the ones most often invoked to justify removing things, and the ones most often misapplied while doing it.

• Hick's Law — decision time rises with the number of options presented at once.
• Miller's Law — about seven items in working memory — and what it is routinely misquoted as saying.
• Occam's Razor — the simplest solution that works, without stripping what people came for.
• Tesler's Law — complexity cannot be removed, only moved — the question is onto whom.

What it costs to actually do the thing

Interaction cost: the physical and temporal price of using an interface, as opposed to understanding it.

• Fitts' Law — target size and distance decide how fast something can be hit.
• The Doherty Threshold — about 400 milliseconds, above which attention starts to leak.
• Postel's Law — be liberal in what you accept — move the burden of precision off the user.

What people expect before they arrive, and remember after they leave

Three principles about everything outside the session itself: the conventions someone brings with them, and what survives once the session ends.

• Jakob's Law — people spend most of their time on other products and expect yours to match.
• The Peak-End Rule — the most intense moment and the ending carry the memory, not the average.
• Parkinson's Law — work expands to fill the time — and so does scope.

The ones about what gets built at all

Two that operate above the interface, on the decision of what to make.

• The Pareto Principle — a small share of features carries most of the value — finding which share is the work.
• Rams' ten principles — read against digital products rather than the industrial design they were written for.

Where the laws stop being useful

Every one of these describes something true about people. None tells you what to build. A principle can rule an option out; it cannot choose between the options that survive.

The failure mode is using a law to end a discussion. “Hick’s Law says fewer options” is not an argument for removing a feature someone depends on — it is an argument for not showing every option at once, which is a different change. The laws are most useful when they name the trade-off, and least useful when they are used to avoid making it.

Common questions