This client with two content heavy websites of over 2000 pages each needed an expert review and user testing for these sites. Because the sites were so large, we set specific investigation goals to specifically focus on information architecture, content quality, and search.
For the heuristic review, we pared down a list of heuristics to focus on structure, content, and search that included (among others):
- The home page orients the user to what the site is about
- Interior pages are well-laid out, have appropriate content for the audience, and are clear about contents in the section
- The information structure is optimized for the main user scenarios
- Navigation choices are ordered in the most logical or task-oriented manner. The terms used for navigation are unambiguous and jargon-free.
For each of these heuristics, we had a ratings scale for the severity of the problem with assessment guidelines. The organization would then be able to run the expert review again using the same ratings scale, create a score for each grouping, come up with an overall score, and see how much their website had improved. (Which they did the following year!)
In the user testing, we created a test plan focused on the main tasks of the site, as defined by the client.
Results
For the heuristic review, we found a number of issues and illustrated them for the client. Some of the issues centred around search filtering, design, and information structure. We illustrated the issue and, where possible, recommended fixes or further investigation.
For this testing, we conducted 24 one-on-one sessions in three cities. We methodically collected and analyzed our results with specific metrics so the client could run the same tests again and judge the website improvements.
One metric we selected was the user satisfaction scale.
For each task, we tracked how easy or difficult it was for a user to complete the task, based on a predefined scale. We then illustrated the success using green, yellow, and red for easy, somewhat difficult, and difficult. Based on these ratings, the client could have a specific number they could measure again to judge the impact of website improvements.
