VKI Usability evaluations are customized to meet the unique requirements of each client. In general, the process involves five steps:
Let's take a closer look at each step:
VKI custom tailors each usability evaluation to the needs of the client. The first task is to decide what needs to be tested. Depending on the type of website being reviewed, the client's objective and the budget, usability tests may be performed on:
Before conducting usability tests, VKI will work with the client to ensure we have all necessary information to conduct the tests intelligently. Depending on the nature of the evaluation, we may ask for detailed traffic reports, customer demographics / psychographics, sales data, etc.
We will not burden you with requests for every conceivable piece of data. In consideration of your privacy – and of the value of your time – we take a practical, common sense approach. We'll only request information that will actually help us conduct the tests and improve the quality of the results.
In preparing a usability evaluation, at least three usability experts will independently examine over 500 aspects of your website. (This is for a site-wide evaluation. The number of factors evaluated will obviously be lower in more restricted tests.) Though it may sound like overkill to have three or more experts review the site, the fact is, there is a subjective nature to several points of investigation. A consensus is necessary before declaring a problem.
After independently reviewing your website, the usability experts will meet and compare notes. Potential problem areas will be reviewed, and where necessary discussed with experts in other fields. For example, if there is a question regarding the appropriate use of title tags, the evaluators would seek the opinion of our in-house Search Engine Optimization (SEO) experts.
The next step is to conduct usability tests on real users. We've found there's simply no substitute for observing target customers attempting to navigate a site. The precise nature and format of the usability tests will vary depending on the nature of the site being evaluated, as will the number of users tested.
Where quantitative data is sought, a large number of test subjects are required to obtain statistically meaningful results. In more open-ended, exploratory evaluations (which represent the bulk of our projects), a surprisingly small number of users are required. In such cases, 75% of usability issues will be revealed after just three or four properly-run tests. And most of the remaining 25% of usability issues will have been uncovered in the expert evaluations. After four tests, the law of diminishing returns usually becomes obvious. To be safe, however, we test a minimum of five users. If the fifth test is still revealing new usability issues, we'll keep running further tests.
The final step is to compile the information gathered during all the testing processes, and to distill and organize it into a readable, actionable document.
