Evaluating navigation information architecture
In software development it is more or less a given that you can never forsee upcoming changes and shifts in techology at the time of design. From this derives the need for continuous adaptation. While this is surely a must, experience shows that unless you’ve had an unchanged team throughout the whole process with one of them having a crystal ball, you’ll end up with tangling code and a growing need to refactor what you did.
When we say experience we mean it as we went through the same in Sense/Net. But now, at the dawn of creating the new version of Sense/Net ECM, we can finally stop and give some thought on how we would re-organize its structure for it to make sense (pun intended) at present.
Beign open-source and reliant on our Community has also been something that would influence the way we approach software development. Therefore, it is only natural that we call on you, our Community to share your ideas and validate our hypoteses when we come to such an important milestone.
We created a tree testing tool to validate our approach on how we will reorganize the sturcture and wanted to give you some insight on how UX development works at Sense/Net. If during the article, you feel like taking part, just drop us an e-mail to firstname.lastname@example.org and we’ll send you the specifics.
The Tree Testing method
Tree-testing is a lesser known UX method but can substantially help with improving problems in navigation. A tree test is similar to a usability test on the skeleton of your navigation with the design "skin" removed. It allows you to isolate problems in findability in your taxonomy, groups or labels that are not caused by issues with design distractions, or helpers.
We also removed the search label from the navigation, since we wanted to isolate the causes of navigation problems to learn about them by forcing users to browse by clicking.
Tree tests are ideally perfomed to:
- Set a baseline "findability" measure before building the navigation and structure. This will reveal what items, groups or labels could use improvement.
- Validate a change: we run this test to measure the effect of the change of IA we made. This helps tell us quantitatively if we've improved findability, or just introduced new problems.
With the Tree Test we will:
- Detect navigational issues early in the design process. It is more effective to get the structure right from the beginning rather than fixing it later down the line.
- Analyze first and all attempts where participants had trouble navigating before your site goes live.
- Effectively measure how well users can find items in a hierarchy
- Collecting real data from users
- Can users successfully find items in the tree
- Can they find them directly, without having to backtrack
- Can they choose between topics quickly
- Which part of the tree is working well and which doesn’t?
If all goes well, the Tree Test will ensure we gain quantifiable data to prove the effectiveness of IA modification.