It seems every time I conduct a usability testing session, I learn something new and the experience, however humbling, leads to a number of important discoveries for the client. Usability testing is valuable because every site is different and clients goals vary, so you really must do it to make sure your site is functioning as desired. However, these recent sessions reminded of some core principles that everyone should know.
I was reminded of several best practices during a recent usability testing session, but the lessons are not specific to the client in question. They are universal and can/should be applied across almost all sites. I am hoping that others will find them useful and that they will serve as evidence of the value of usability testing, both in terms of revealing the predictable and unpredictable.
Lesson 1: Lead with your most important content
During the testing sessions, users quickly scanned the pages and stopped when they arrived at a data table or visualization. In several instances, the users would arrive at a page that contained the content they were looking for in a visualization that was more than one to two screen lengths below the original screen. Unfortunately, they would click away from the screen prior to reaching the visualization. While it almost seems obvious, it is important to remember that your most important content on each page should be featured top-center. This principle closely aligns with what journalists know as the inverse pyramid writing style.
Lesson 2: Design search engines to search the entire site
Users often relied on the global search to complete their tasks. They expected that this search would search all of the web site's content. Unfortunately, the site contained several large data sets that were omitted from the global search results. In consequence, users had difficulty finding that information. The Research-Based Web Design & Usability Guidelines provide the follow advice:
Lesson 3: Ensure that iconography is hyper-linked
During one of the sessions, a user became frustrated and clicked on the "Play" icon next to a link that read, "Need help? Watch the screencast." The user clicked this icon several times, but never clicked the underlined text just to the right of the icon. The end result was that he abandoned the task and became frustrated with the system. As it turns out, most of the icons on the web site were hyperlinked, but this particular icon had a bug in the code. Nonethless, this best practice cannot be overlooked.
Lesson 4: Establish credibility
Some organizations have a difficult time establishing themselves as a credible source. This client, however, is not one of them. They are taking all of the necessary steps to ensure that they are perceived as a credible resource. Their site does an excellent job of establishing its credibility by:
- Providing articles that contain citations and references
- Including a useful set of frequently asked questions and answers
- Providing links to outsider resources and material
- Ensuring that the site has a professional, polished visual design
- Providing links to source data sets, citing .gov domains in particular
During several of the sessions, the participant would point to the data source and acknowledge that it is good that the client was using government data.
Lesson 5: Ensure visualizations/quantitative information is easily understood
During our usability testing sessions, the users would arrive at a data table or visualization and would quickly process the information it contained without reading the supporting text. They expected the visualization to communicate enough information such that they should not have to do much additional reading. This expectation lead them to ignore the text and ingest only the visualized data. To accommodate this expectation and help users quickly process visualizations / data tables, your site can:
- Label the visualization in a way that helps the user draw a conclusion
- Where possible, include the visualization's label, controls, and the visualization all in one view
- Place the image's legend and controls close to each other so that users can more readily interpret how the visualization changes when a change is applied
- Include explanatory text below the visualizations
- If the data is particularly difficult to interpret, consider adding a contextual help link near the visualization's header that explains why analyzing this data presents certain challenges
Conclusion
In retrospect, these revelations seemed obvious and intuitive. How could they have been overlooked? This points to the need to conduct usability testing, even if only to reveal certain findings that you already knew, but needed to remember. And, in most cases, usability testing does more than reveal the known — it uncovers seemingly unpredictable responses and experiences, many of which are specific to your audience and how they are interacting with your site and its content, that are dramatically affecting your site's user engagement and effectiveness.
Forum One News
Brian grew up on land that used to be part of his grandfather's dairy farm and spent hours exploring the outdoors. That kind of roaming and curiosity is a great foundation for an information...





Comments
K.I.S.S.
Wonderful info. I think many of us forget about the K.I.S.S. (Keep It Simple Stupid) method of doing things. It's really difficult sometimes to put yourself outside the box and view what you're doing from the eyes of someone who's not as informed and therefore, is not sure how to get from point A to B effectively. I've found, like you, that starting with starting with our most important content in a broad scope and macro sense helps to get the ball rolling. If the information isn't what they're seeking and they do a search after, it usually helps. Bombarding them right off the bat is usually a turn-off and leads them to search elsewhere for their answers!
:)
Hanna
FreedomIQ
business phone systems
RE: Great Insights
Post new comment