During the production of a new website for MySight York, we organised a live, in-person user testing session to understand how easy the website was to use for the charity’s key stakeholders: those living with partial or full sight loss.
Over the course of three hours, a small group of four participants took up the challenge of completing specific user journeys before sharing their feedback with us 1:1 and in a group discussion.
In this article we’re going to share a little more about how we ran this session, our key findings, and what we did to improve MySight York’s new website before launch.
How we conducted user testing
For website users who carry additional needs – whether they be physical, cognitive or emotional – it’s not only important to understand these needs, but to go the extra mile to cater for them. From day one of the MySight York project, we knew that partially sighted/blind users were a key stakeholder group, so this project was designed from the ground up with that in mind.
But user testing proved that what we had done was still not quite enough. Here’s how we uncovered problems in what we were building.
1. We tested the website, not the user
To earn the trust of test participants, we stressed that it was the website being tested, not them. Many people – particularly those who describe themselves as a not “tech savvy” – are quick to blame themselves when something goes wrong. We are big believers that technology should support user aims, and that in most cases, the technology could do better to meet user needs.
2. We provided destinations, not pathways
Critically, our testing involved goal-oriented tasks. We gave our participants challenges such as “Imagine you’ve recently been told you have glaucoma. You don’t know much about it yet. Show me how you’d find out what MySight York could do to help, and find one service that might help.” These types of challenges give the user a destination to reach without prescribing a path to get there. As you’ll see in the findings below, this was particularly illuminating.
3. We helped only when necessary
To thoroughly test a website or app, participants must be allowed to “struggle”. Whilst it can be uncomfortable to observe a person getting stuck with the task you gave them, on a website you designed and built, it is a critical part of the process. That being said, in order to complete the task and discuss it, we occasionally needed to help our participants reach their destinations.
The key takeaway from user testing, however, may not have even been in the testing itself, but in the willingness of our participants to be so open and helpful. Many pieces of thoughtful feedback came in the form of lateral ideas; unrelated directly to the tasks at hand, but in running commentary and the post-testing discussion. Without this willingness to engage, our testing would have been far less effective. Here’s what we discovered.
For website users who carry additional needs - whether they be physical, cognitive or emotional - it's not only important to understand these needs, but to go the extra mile to cater for them.
What we discovered in user testing
The user testing session lasted three hours and finished in a group discussion about the overall experience, expectations of the new website, and thoughts on living with sight loss more generally. We collated our findings and presented them back to the team at MySight York. What struck us the most was that our assumptions – especially those around best practices – were thoroughly tested. For example, our rationale for not including a website search function had been because users either enter on their intended page from Google, or browse directly to it through clear navigation. This was just one erroneous assumption on our part, and one that we fixed as a result of testing.
Content density
It was clear through observation that both partially-sighted and blind test participants struggled with the amount of content on the pages. All participants naturally accessed the mobile version of the website, even on a large laptop screen due to zooming in. This meant the scroll went on and on. We addressed this by reducing and consolidating content, reviewing page hierarchy to place important content higher up, and introducing a “back to top” link for quick return to the top of the page.
Inaccessible accessibility features
Somewhat ironically, our accessibility toolbar was hard to find and use by all participants. And once located and opened, it was also unclear how to close again. We addressed this by increasing font size, increasing colour contrasting, and adding a close button.
Confusing page titles
It was noted by the participants that they could not easily locate the content they were being asked to find. This, in part, was due to content density (covered a moment ago), but it also spoke to confusing page titles. We had originally planned verb-based navigation such as “living well with sight loss” and “get support”, but after discussion these were simplified to “sight loss information” and “support & activities”. We listened to how both MySight York staff and their stakeholders actually refer to these functions – then updated the website accordingly. We also added a site-wide search function as another route in.
Interactive elements
We discovered that participants would confuse our interactive buttons and card links with labels and content blocks that were not intended to be clicked or tapped. The visual styling was not distinct enough between the two. Our solution was to redesign these elements so that non-interactive elements did not appear so.
Lack of content warmth
We partnered with a copywriter living with sight-loss to deliver content for the website, but some aspects of the site didn’t quite connect with service users as intended. One of the test participants suggested the content could be more empathetic and reassuring, with references to real life experiences, such as “finding your way around the kitchen” or “struggling to use the cooker”. We reviewed key engagement areas of the site and introduced more empathetic, real-life examples where appropriate.
Many of these changes were minor, but made for a large overall improvement to the user experience of the website – particularly for those living with sight loss.
Running a user testing session
If your charity would benefit from engaging real-life users to test your website or app, we hope that this case study will give you the confidence and a basic structure to get started. However, if this is something you’d like support with, please feel free to contact us to find out how we could help.