011 min
The problem UX principles solve
Picture a team that has just built a refund screen. In the design review, a product manager says the screen feels confusing. The designer thinks the layout is fine. The engineer thinks the labels are fine. The team argues about taste, and nothing changes.
A principle turns that feeling into a question with an answer. After a person presses the refund button, can they tell that the refund began? Does the same word mean the same thing on every screen? Can they stop before the money moves? Each question points at something the team can check, fix and check again.
That is the job of UX principles. UX means user experience: everything a person meets when they use a product, from the first screen to the help page. The principles give a team shared questions in place of private opinions.
021 min
Where UX principles come from
Nobody wrote one official list. Several people wrote their own in different years, and the lists overlap.
In 1985, John Gould and Clayton Lewis recommended three steps for designers: study users and their tasks early, measure real use, and improve the design in rounds. Ben Shneiderman, a computer scientist at the University of Maryland, wrote eight rules for interface design, dating the first version to 1985. In 1988, Don Norman published a book on how people deal with everyday objects. It was originally titled The Psychology of Everyday Things and is now called The Design of Everyday Things. Jakob Nielsen and Rolf Molich developed the first set of usability heuristics in 1990. Nielsen revised that set in 1994, and the ten heuristics people quote today are the revision.
Later work widened the view from the screen to the person's whole dealings with a product. The standard ISO 9241-210, published in 2010, set out requirements and recommendations for human-centered design across the life of an interactive system. Companies and governments then began to write principles of their own.
031 min
UX principles vs. heuristics, guidelines and laws
Designers use four words for this kind of advice, and the words overlap. The table separates them by how specific and how well tested each kind is.
| Kind | What it is | Example |
|---|---|---|
| Principle | A broad statement of what a good design does | Tell the person what happened after every action |
| Heuristic | A principle packaged as a checklist item for reviewing a design | Nielsen's ten, used in heuristic evaluation |
| Guideline | A concrete instruction that can be checked exactly | A minimum size for a button in WCAG 2.2 |
| Law | A relationship measured in experiments | Fitts's law: time to reach a target depends on its distance and size |
Most UX principles sit in the first two rows. They are conventions that practitioners agreed on after seeing the same problems many times, not results from one experiment. Nielsen's group describes its heuristics as broad rules, not detailed guidelines. Presenting a convention as a research result gives it more authority than it has.
One principle, consistency, rests on a law-like observation. Jakob's law says people expect a new product to work like the ones they already know. So a product is consistent in two ways: with itself and with the products around it.
043 min
The UX principles that appear in most lists
Lists differ in length. Nielsen has ten heuristics and Shneiderman has eight rules. Peter Morville's honeycomb, covered below, has seven qualities. The Decision Lab says there is no fixed number. The same ideas return in most lists, though. The table gives the common ones and the question each settles.
| Principle | The question it settles |
|---|---|
| User-centered design | Do we know what the people who use this need, and have we checked? |
| Usability | Can people learn it, work fast once they know it, remember it, and recover from mistakes? |
| Consistency | Does one word or control mean one thing everywhere, and does it match other products? |
| Feedback | After an action, can a person tell what the product did? See visibility of system status. |
| Error prevention | Can the design stop a serious mistake before it happens? |
| User control | Can a person undo an action or leave a step? |
| Recognition over recall | Does the person have to remember something the screen could show? |
| Visual hierarchy | Does the eye go to the most important thing first? |
| Simplicity and plain language | Is every element needed, and are the words ones the person already uses? |
| Flexibility | Does the product serve both a beginner and an expert? |
| Context | Does it suit the place, the device and the task? |
| Accessibility | Can people with different abilities and settings use it? |
Some principles are about structure. Navigation and information architecture decide whether people can find things. A person who opens a new product has a guess about how it works, called a mental model, and clear structure and steady labels protect that guess. Progressive disclosure shows advanced options only when a person asks for them, so a beginner is not overloaded and an expert still reaches everything. Whitespace and plain words keep a screen readable, because each extra element competes with the ones that matter.
Others are about how a team works. Start from what users need. Test with real users early and again after each change. Record the answers in a design system, a shared set of components and rules, so that consistency does not depend on anyone's memory. A refund button built from that set behaves like every other button.
Accessibility has a measurable standard. WCAG 2.2 success criterion 2.5.8, Target Size (Minimum), asks for targets of at least 24 by 24 CSS pixels, or enough space around smaller ones. People with hand tremors or limited fine movement cannot press small controls that sit close together.
Two kinds of error
Error prevention covers two different failures. Nielsen's group separates slips, which are unconscious errors caused by inattention, from mistakes, which are conscious errors that come from a mismatch between the person's mental model and the design. A confirmation step helps with slips. Clearer labels and structure help with mistakes.
051 min
Why UX principles work
A principle is useful when one question covers many problems seen before. Nielsen's history shows how. In 1994 Nielsen revised the heuristics after a factor analysis of 249 usability problems, a statistical method that groups related items. The ten heuristics were tested against a body of observed problems, not picked by taste alone.
The two gaps behind many principles
Norman gives a second reason. A person using any device faces two gaps. The first is the gulf of execution: the person cannot tell which actions are possible or how to perform them. The second is the gulf of evaluation: the person cannot tell what the system did after the action. Norman's book names four ideas for reducing them: visibility, a good conceptual model (the person's picture of how the thing works), good mappings (a clear link between each control and its effect), and feedback.
The second gap is the larger risk on a screen. The gulf of evaluation is the effort a person needs to work out what state the system is in and whether their action did what they wanted. Feedback reduces it. That is why so many lists begin with it.
062 min
UX principles applied and broken
Return to the refund team. They ask their three questions. The screen shows a spinner and then goes back to the order list, so nobody can tell that the refund began. The team adds a message that states the amount and when it will arrive, and a link to cancel while the request is pending. One change gives feedback and the other gives control.
Shneiderman's own example of closure is an online shop that moves a buyer from choosing products to checkout and ends on a clear confirmation page. Closure means a group of actions has a beginning, a middle and an end, and the end tells the person they are done.
A product that broke the rules
Real products show what happens when a principle is ignored. In November 2012 Nielsen Norman Group published a test of Windows 8, then new. It invited 12 experienced PC users to try the system on regular computers and on Microsoft's Surface tablet. The new look used flat tiles and flat text with few cues to show what could be clicked. Many testers did not click Change PC settings, because it looked more like a label than a command.
The problem was visibility. The command was there, but nothing showed that it could be pressed. The cue that shows what an object can do is called an affordance, and Norman's work made the term common among designers.
Principles written for one organisation
Governments and companies also write their own short lists. The UK government's design principles begin with start with user needs, and include do the hard work to make it simple, and understand context. The ninth says to be consistent, not uniform. Facebook shared a list in 2009, and Nokia announced one with its Pure design system in 2023. A list like this is useful because it is short enough for a team to remember.
071 min
UX principles beyond the interface
Usability is only one part of user experience. Peter Morville, an information architect, drew a honeycomb of seven qualities. His aim was to help clients see why they must look past usability. A product should be useful, usable, desirable, findable, accessible, credible and valuable.
In his honeycomb, usability is one quality among seven, and Morville calls it necessary but not sufficient. He also says a team should decide the trade-offs between the qualities openly, since the right balance depends on the product, its content and its users.
UX is wider than the interface
Norman and Nielsen make the same point with a site of movie reviews. On a site of movie reviews, the screen for finding a film can be perfect and the experience can still be poor for someone who wants an independent release, if the database lists only major studios. Interface principles cannot repair a gap in what the product offers. That gap needs a different kind of work, such as research into what people want.
Usable is not the same as useful
Nielsen separates usability, which is how easy the functions are to use, from utility, which is whether the product has the functions people need. A product needs both usability and utility, since an easy tool that lacks the needed function fails the person as surely as a capable tool that cannot be operated.
081 min
Where UX principles mislead
Because a principle is compressed experience, it can be wrong in a setting the original authors never saw. Shneiderman says his rules came from experience and must be checked and adjusted for each area of design. He adds that no such list can be complete.
Consistency does not mean sameness
A team that applies consistency as a rule can keep a weak pattern unchanged. The UK government's own principles say their patterns are not a rule book. Teams share what works, and a pattern can change when someone finds a better way or when user needs change.
Culture and novelty
The Decision Lab points to two more limits. What works well in one culture may not suit another, so a team should adapt the principles. And a team that follows the principles too closely may produce only familiar designs. This is the cost of Jakob's law: familiar is easier to use, and it is also harder to improve.
Knowing a principle is not following it
Gould and Lewis surveyed designers in their 1985 work. Their answers suggested that some designers believe they follow the advice when in fact they do not. A team can recite the list and still ship a screen that breaks it. Only a check against the real screen, and then against real users, shows which one has happened.
091 min
Applying UX principles
These moves fit a team this week.
- Choose a short list. Start from a published set such as Nielsen's ten, then cut or reword each item to fit the product.
- Write each principle as a question that a reviewer can answer yes or no for one screen.
- Review the main task screen by screen, and record each answer.
- Test with real users doing real tasks. Nielsen Norman Group advises testing early and repeatedly, with representative users.
- Put the fixes into the design system, so the next screen starts from the corrected version.
Stop when the list starts to make the product worse. If testers struggle with a design that passes every question, trust the testers and change the list.
101 min
Frequently asked questions about UX principles
What is the most important UX principle? No source ranks them. The UK government puts start with user needs first, and Gould and Lewis began with early focus on users. The right order depends on the product, so state the trade-offs openly.
Are UX principles only for websites? No. Norman drew his ideas from everyday objects such as doors and projectors. Shneiderman says his rules are a good starting point for mobile, desktop and web designers.
Do UX principles replace user research? No. Principles predict common problems, but only testing shows the problems in your own product. Nielsen Norman Group recommends early, repeated tests with real users.
Do UX principles go out of date? The core lasts. Nielsen published his heuristics in 1994, and his group reviewed the page again in January 2024, updating the names, descriptions and examples.
?4 questions
Questions people ask
What is the most important UX principle?
Are UX principles only for websites?
Do UX principles replace user research?
Do UX principles go out of date?
§10 sources
Sources
John D. Gould and Clayton Lewis, "Designing for Usability: Key Principles and What Designers Think", Communications of the ACM 28(3), March 1985.
Ben Shneiderman, "The Eight Golden Rules of Interface Design", University of Maryland.
Jakob Nielsen, "10 Usability Heuristics for User Interface Design", Nielsen Norman Group, 1994, last reviewed 30 January 2024.
Wikipedia, "The Design of Everyday Things" (on Don Norman’s 1988 book, revised 2013).
Show all 10 sourcesShow fewer sources
Jakob Nielsen, "Windows 8 — Disappointing Usability for Both Novice and Power Users", Nielsen Norman Group, 19 November 2012.
Don Norman and Jakob Nielsen, "The Definition of User Experience (UX)", Nielsen Norman Group.
Peter Morville, "User Experience Design", Semantic Studios.
Government Digital Service, "Government Design Principles", GOV.UK.
W3C, "Understanding Success Criterion 2.5.8: Target Size (Minimum)", WCAG 2.2.
The Decision Lab, "UX Principles".
