A brief introduction to Heuristic Evaluation

Heuristic evaluation was developed by Nielsen and Molich in 1989 as a cheap and cost-effective evaluation method of evaluating the user interfaces of computer systems. It is essentially an informal and somewhat subjective method of usability analysis. Typically it involves a set of evaluators examining the user interface of an interactive product and using their knowledge of the intended users and tasks to predict problems. The evaluators judge the compliance of the interface with recognized usability principles (known as heuristics).

Heuristic evaluation can be applied at any stage in the development lifecycle once the user interface of a system has been designed, as long as there is somethng that
describes the user interface - from a storyboard or prototype to a fully working system.

What exactly are heuristics?
Essentially a heuristic is a guideline (eg "is feedback provided at all stages?") to be used as part of a checklist, when assessing an interface. For example I would say "yes, feedback is usually provided for the action of submitting a form on an ecommerce website". In contract feedback is often not provided for certain cell phones that are running out of credit (on Pay-as-you-go systems); the result being that you are cut off before you can do anything about it.

Our use of the term ‘heuristic’ on this website is as a specific application of a principle and takes the form of a question to be applied when evaluating an electronic product. As there are so many of them (about a hundred), we have placed them into various categories of like-minded heuristics (eg those related to "system status").

Heuristic evaluation is very cost-effective. It is quick, intuitive and simple to administer and requires only a small number of evaluators (5 should be sufficient).

It does not involve real users and is not repeatable (you never get the same set of problems twice). It is subject to evaluator bias and is also extremely dependent on the skill of the evaluator.

For more details please refer to Chapter 13 of the book "Interaction Design: Beyond Human-Computer Interaction".

When would I use it?
Heuristic evaluation is useful for doing a first pass at assessing the usability of a new product. Suppose you were developing a novel application for a palm device. Before exposing your design to real users, you could identify many potential problems with the interface, by first conducting a quick heuristic evaluation. However, there are hundreds of heuristics you could use to evaluate an interface, so how do you decide which ones to use?

Why use this interactive tool?
The aim of this tool is to help you with the dilemma of choosing appropriate heuristics. We provide you with a 'hands-on' exercise to think about which heuristics are appropriate for different devices. We also provide you with immediate feedback to enable you to compare your suggestions with the previous selections of experts. At the end we hope you reflect on how and why your selections differ (if they do).

Back to heuristics