Usability heuristics are Jakob Nielsen’s 10 general principles for interaction design, called heuristics because they are broad rules of thumb rather than specific guidelines, and they work as a checklist for judging any website.
Because they are broad, the principles apply equally to a single layout from a template collection, as in the guide to web design templates, and to a fully built store.
Where the heuristics come from
Nielsen Norman Group publishes the current wording. Jakob Nielsen originally developed the heuristics for heuristic evaluation in collaboration with Rolf Molich in 1990, and later refined them using a factor analysis of 249 usability problems, aiming for a set with maximum explanatory power. NN/g updated the article in 2020, adding more explanation, examples and related links, and the names below follow that published list.
The word “heuristic” matters. These are not laws and not pass-or-fail tests. They describe patterns that, when broken, tend to produce confusing interfaces. Applying them takes judgment about the audience and the task.
Status, language and escape routes
Visibility of system status. The design should always keep users informed about what is going on, through appropriate feedback within a reasonable amount of time. On a website, this means a loading indicator on a slow search, a clear confirmation after a form is sent, and a visible current state in navigation and checkout steps. NN/g’s tip is to present feedback as quickly as possible, ideally immediately.
Match between the system and the real world. The design should speak the users’ language, using words, phrases and concepts familiar to the user rather than internal jargon. A menu labeled with the company’s internal department names fails here; a menu labeled with the tasks visitors came to do passes.
User control and freedom. Users often perform actions by mistake and need a clearly marked emergency exit to leave the unwanted action without going through an extended process. Practical forms of this include a Cancel button on a dialog, an undo link after deleting an item, and a back button that behaves as expected in a multi-step flow.
Consistency, prevention and memory
Consistency and standards. Users should not have to wonder whether different words, situations or actions mean the same thing. NN/g distinguishes internal consistency within a product from external consistency with established industry conventions, and cites Jakob’s Law: people spend most of their time using digital products other than yours, so those products set their expectations. A logo that links to the home page and a shopping cart in the top corner are examples of conventions worth keeping.
Error prevention. Good error messages matter, but the best designs carefully prevent problems from occurring in the first place. NN/g separates slips, which are unconscious errors caused by inattention, from mistakes, which are conscious errors based on a mismatch between the user’s mental model and the design. Helpful constraints and good defaults reduce slips; confirmation steps and clear warnings reduce mistakes.
Recognition rather than recall. Minimize the user’s memory load by making elements, actions and options visible. Information needed to use the design, such as field labels or menu items, should be visible or easily retrievable when needed. A placeholder that disappears as soon as someone starts typing forces recall; a persistent label does not. The same thinking drives sound web form design, where labels, hints and error text stay in view while a person fills in each field.
Efficiency, focus, recovery and help
Flexibility and efficiency of use. Shortcuts, hidden from novice users, may speed up the interaction for the expert so that the design caters to both inexperienced and experienced users. Examples include keyboard shortcuts, saved addresses at checkout and customizable dashboards.
Aesthetic and minimalist design. Interfaces should not contain information that is irrelevant or rarely needed, because every extra unit of information competes with the relevant units and diminishes their relative visibility. NN/g adds that this does not require flat design; it is about keeping content and visual design focused on the essentials.
Help users recognize, diagnose and recover from errors. Error messages should be expressed in plain language with no error codes, precisely indicate the problem, and constructively suggest a solution. “Something went wrong” fails all three. “The email address needs an @ sign” passes.
Help and documentation. It is best if the system does not need additional explanation, but documentation may be necessary. Help content should be easy to search and focused on the user’s task, concise, and list concrete steps. Where possible, it appears in context at the moment the user needs it.
The heuristics at a glance
| Heuristic | Typical website check |
|---|---|
| Visibility of system status | Does the page confirm that an action worked? |
| Match with the real world | Do labels use visitors’ words? |
| User control and freedom | Can every step be undone or cancelled? |
| Consistency and standards | Do similar elements look and behave alike? |
| Error prevention | Do defaults and constraints stop likely slips? |
| Recognition rather than recall | Are options visible, not remembered? |
| Flexibility and efficiency | Are there shortcuts for frequent tasks? |
| Aesthetic and minimalist design | Is anything on the page competing with the main task? |
| Error recovery | Do messages say what went wrong and how to fix it? |
| Help and documentation | Is help searchable and shown in context? |
Heuristic names and descriptions are summarized from the Nielsen Norman Group article 10 Usability Heuristics for User Interface Design.
Running a heuristic evaluation
A heuristic evaluation is a method for identifying design problems in a user interface, in which evaluators judge the design against a set of guidelines that make systems easy to use. NN/g recommends Jakob Nielsen’s heuristics for assessing usability, and says the method suits glaring problems and can be especially helpful early in the design process, even on prototypes.
The same article describes the working steps. Prepare by narrowing the scope to one task at a time, one section of the site or app, one user group or one device type. Then each team member evaluates the interface independently, and NN/g suggests timeboxing this to about 1–2 hours. Finally the team consolidates the identified issues.
The method has a limit that NN/g states directly: heuristic evaluations cannot replace user research. They help find likely issues without testing with participants, which stretches a limited research budget, but real users still need to try the design.
Applying the heuristics before and after launch
The checklist works best at the layout stage, when changes are cheap. Reviewing a website wireframe against the principles catches missing confirmations, unclear labels and absent exits before any visual design exists. After launch, the same list gives reviewers a shared vocabulary: a report that says “this checkout step breaks error prevention” is easier to act on than “this feels wrong”.
Priorities differ by site. A shop might weight error prevention and recovery most heavily, a documentation site might weight help and recognition, and a news site might weight minimalism and consistency. NN/g’s own advice to prevent high-cost errors first, then little frustrations, is a sensible way to order the fixes that come out of any review.
Frequently asked questions
Are usability heuristics the same as usability guidelines?
No. They are broad rules of thumb, which is why the term heuristic is used, while guidelines are specific instructions for a given element or situation.
Who created the heuristics?
Jakob Nielsen developed them with Rolf Molich for heuristic evaluation and later refined the set. Nielsen Norman Group maintains the published version.
Can heuristic evaluation replace user testing?
No. It helps find likely problems cheaply, but evaluators are not the audience, so testing with real users is still needed.
Which heuristic matters most?
None is ranked above the rest. The right priority depends on the site: prevent high-cost problems first, then smaller frustrations.
