National Museums Scotland – putting users first

Creating a new website often involves lots of stakeholders, workshops, and interviews. When working with National Museums Scotland on their new website, as well as having internal meetings, everyone was keen to put users first. So user testing was built-in from the start.

During this project, we collaborated with the teams at One Further and National Museums Scotland to run various rounds of user testing.

Thirteen tests were created in total, with more than 900 users taking part. Tests were a mix of moderated and unmoderated tests, usability tests, A/B tests, preference tests, 5 seconds tests, and surveys.

Making user testing part of the process

Most user testing took place during the Design phase of the new National Museums Scotland website. At this point in the project, we were keen to test ideas around site structure, menu navigation, language, and top-level landing pages.

User testing can take time, so throughout the Design process, we planned-in time to include user testing. This often meant building prototypes for testing and, while tests were taking place, working on other aspects of the Design phase.

Navigation was a key focus. To test this we designed and built prototypes that One Further then used for testing. The process was highly collaborative and open, with everyone coming to the project without preconceived ideas.

Thanks to Alix Geddes (One Further) and Susan McAteer (National Museums Scotland) for sharing their experiences and the things they found most surprising during user testing:

Getting navigation right

Alix Geddes – Senior Content Analyst, One Further

One Further conducted research for National Museums Scotland, evaluating the website’s usability, navigation, and menu language. To gather user feedback, One Further used moderated and unmoderated testing methods.

Supercool developed 3 menu prototypes for testing: a traditional desktop menu, a burger-style desktop menu, and a mobile menu. A large cohort of testers was tasked with completing various tasks in an unmoderated setting, while individual interviews were conducted with a smaller selection of chosen participants. These insights were combined to provide insights into each menu design.

A month later, a second round of testing was carried out, incorporating learnings from the initial tests. This time, only one refined prototype was tested thanks to Lyssna’s diverse range of testing options. Supercool’s use of Figma to design the prototypes made it easy to carry out the tests with a wide range of participants.

In this project, the most interesting finding emerged in the Collections section of the website. Participants were confused by all the numbers displayed when searching for objects i.e. how many objects are in the collection, the number of search results, the number of objects that have been digitised. This made it difficult for them to grasp the scope of the collection, or the extent of their search.

This confusion created a significant barrier, preventing users from finding what they were looking for. Further rounds of testing were conducted to drill down into this issue and uncover useful data for the website team to take forward, i.e. reducing the number of figures appearing on the page, and clearly explaining what each means.

Sharing rich content

Susan McAteer – UX Content Producer, National Museums Scotland

On our previous website there was a section called 'Explore'. It featured articles, films, and resources that expanded the stories of the museum’s collection. This section had become a bit muddled over time, featuring articles that were quite academic in formats that were tricky to follow, as well as resources for schools and families.

We wanted to refine this section to present engaging stories related to the collection, written and presented with particular user groups / use cases in mind. And as part of this rethink, we reconsidered what this section of the website should be called.

Across the sector there is no one way these sections are defined. Looking at other organisations, the title for these sections varied widely – with examples including Magazine, Articles, Blog, 'Explore and Learn', and ‘Read, watch, listen’. The most common name we found was Stories.

We started to use Stories as the default name for this section. It was popular with the digital team and we felt that this represented what the section would be. However, when we tested the name only 10% of users (out of 100 participants) selected Stories as the title for this section; Discover was the clear winner.

Before committing to this title we ran an A/B test to scope where to place Discover in the navigation. Again, the results surprised us in that both the options presented had roughly the same success rate – with 72% and 74% of users selecting the happy path.

These test results allowed us to make a user-first decision. Rather than leading with internal preference or following sector trends, we demonstrated evidence based thinking which was particularly helpful when explaining the changes to colleagues.

Beyond the surprising results that user testing usually presents, I always enjoy talking to users face-to-face. For one usability test, a colleague and I set up a testing table in the museum. Talking to visitors, we gathered other nuggets of information about how they used our website – and why. This is the kind of feedback that would have been impossible to gather through a screen.

Embedding a user-first approach

The digital team at National Museums Scotland successfully put users first when making decisions about the new website. But more importantly, this process has helped to embed a user-first approach across the organisation. As Susan says, this evidence-based thinking is useful when communicating across the organisation. And it's an approach that can continue as the website develops over time.

For more insights like this – plus invitations to our FREE webinar series – join the Supercool Mailing List