A heuristic evaluation is a process by which an expert evaluates a user interface or similar system using a list of guidelines. This is not the same as a user evaluation or usability test in which users actually try out the interface. Instead, a predetermined list of features or aspects of a user interface that are commonly accepted as being beneficial is used to evaluate the interface. A heuristic evaluation is typically faster and less expensive than a usability test, though it does have weaknesses and should be used early in development.
There are different ways in which a heuristic evaluation can be conducted, but it typically begins with a list of criteria or features expected of a strong user interface. This list can come from a number of sources, though the first such basic list was created by Jakob Nielsen and establishes 10 principle design elements that should be included in an interface. Different experts in usability and design can create their own lists, or use these 10 as a starting point for more detailed checklists. When that expert is called upon to perform a heuristic evaluation, then he or she uses the checklist to consider the strengths and weaknesses of a system.
A heuristic evaluation is usually conducted by an expert in usability features and interface design, rather than actual test users. The expert looks at the different elements of an interface and evaluates each part of it according to the checklist he or she has created. This can include the use of “yes” or “no” answers to evaluate if certain elements are present in the interface, as well as a numerical scale to indicate the severity of problems or issues found in the heuristic evaluation. The scale allows program developers to easily recognize the nature of a problem and quickly determine if the resources are available to correct it prior to software release.
One of the major weaknesses of a heuristic evaluation is that it applies common standards to different types of systems. A feature that may be required in one type of software may be unnecessary in another; while some features that might be considered poor design for some programs can actually be beneficial in others. Many companies still utilize experts to perform a heuristic evaluation, however, since the process is faster and cheaper than long-term usability testing using large groups of users. Heuristic evaluations are still beneficial, but they should be used early on in the design and development process so that changes suggested by the evaluation can be considered prior to usability testing that often demonstrates the reality of interface usability.
