Improving information architecture of a community health service website
Tree testing analysis - Melbourne, Australia
Background
I was engaged by Nomat to analyze a tree testing study as part of improving the information architecture (IA) of a community health service website.
About tree testing
Using a software platform called Optimal Workshop, participants were given 12 tasks that asked them to locate some information.
For each task participants clicked through a link ‘tree’ and indicated where they expected to find the information.
More than one solution can exist for each task, and task order was randomized for each participant to account for increasing familiarity with the tree structure as tasks progress.
Research objectives
- Understand how effective the new structure is at supporting the findability of priority information relevant to users’ top tasks.
- Investigate where participants look for information to inform further refinements to the IA structure.
- Identify any issues or concerns with the key changes to the structures.
Impact
Overall, the task success rate of the new IA structure is x% which means x% of participants chose a correct answer across all of the tasks they attempted. This result indicates that the tree is [insight], but may need revisions to help the users achieve the tasks faster and easier.
The study also highlighted IA opportunities to improve the current structures, for example:
- Although [item A] attracted traffic for the right tasks ([task], [task], and [task]) it also attracted numerous unwanted clicks, presumably because the name is too general. Recommendations: Remove [item A] from the top-level sections, and move it to [item B].
- Although [item A] and [item B] were selected by many participants looking for [task], top-level sections were confusing. Recommendations: Relocate [item A] to the top-level categories.
- [item A] and [item B] were not expected by many participants looking for [task]. Recommendations: Relocate [item C] on its own and cross-link it to [item D].
The study revealed some other opportunities to improve the IA. However, due to confidentiality reasons, I can’t show the insights and opportunities here. Feel free to contact me if you want to get to know more about this project.
