In my most recent position, I was in charge of posting internal resources to our company intranet. After hearing about numerous issues that users had with the site, I decided to research the problem in order to propose solutions that would improve the users' experience.
Team: self-led project
Timeline: a few weeks
- What are the biggest pain points for site users?
- How do users categorize information on the site?
Time limitations: because this was a side project at work, I had a limited number of hours to devote to it
Resource limitations: because it was not a formal project, there was no budget
- User interviews/shadowing
- Card sort
- User testing
chose these methodologies based on the fact that they were easy and relatively quick to prepare and conduct. Because I didn't have much time, I used four subjects and did each of the three activities with each.
- User interview - I spoke with the user for 15 minutes to get a sense of how they use technology at work in general, how and for what they used the intranet, and their general impressions of the intranet site (helpful or a pain?).
- Card sort - next, I had users group cards that listed the key items on the intranet homepage into categories, to get a sense of how they would organize the information on that page.
- User testing - last, I had users perform a set of tasks designed to see if they knew how to search for resources they needed, observe how they navigated around the site, and whether they knew where specific key resources were located.
I uncovered a few key insights as a result of the research:
- Site navigation was impeded by the lack of navigation links
- Recommendation: add breadcrumb links and back button on each page
- Information on the homepage was not organized optimally
- Recommendation: reorganize info based on user role rather than topic
I left before I could implement the recommended changes to the site.
If I had been able to make changes, I would have tracked the following metrics, comparing before & after the change:
- # of times key resources were downloaded
- # of unique visitors
The generalizability of the results is limited based on the small number of participants.
In the future, I would expand the number of participants, and would do follow-up tests after proposed changes were implemented.