Four audiences, three devices, two days: testing Middlesex University’s redesigned website
Middlesex University had just redesigned its website and needed to know it worked for everyone it serves. Box UK tested it with four audiences on three devices, and gave it a benchmark and a prioritised list of fixes.
In short
Middlesex University had recently redesigned its website, which has to serve prospective students, the parents and teachers who influence them, and academics. It wanted a health check: how well the new site worked for each of those audiences, on the devices they actually use.
- Box UK tested with prospective undergraduates and postgraduates, parents and teachers, and academics.
- Participants used desktop, tablet and mobile in the same proportions as the university’s own visitors.
- Everyone completed core tasks, plus tasks specific to their group, such as finding a course or comparing accommodation.
- The report gave task success rates and recommendations ranked by urgency and impact.
- A System Usability Scale score gave the university a benchmark for measuring future changes.
Customer background
Middlesex University, based in north London, has around 24,000 students. Like most universities, its website has to serve several very different audiences: prospective undergraduates and postgraduates, parents and teachers who influence their choices, current students, and academics and researchers.
The challenge
Middlesex had recently redesigned its website and wanted an overall health check: first impressions, efficiency, ease of use and satisfaction, for each of its key audiences and on the devices they actually use.
It needed confidence that the redesign was working, and a clear view of where to improve next.
Our approach
We agreed a test plan with Middlesex covering four audiences:
- prospective undergraduates
- prospective postgraduates
- student influencers, such as parents and teachers
- academics and researchers
Recruitment mattered. Prospective undergraduates, for example, hadn’t formally or mentally chosen a university yet, and the other groups included a good mix of backgrounds and specialisms.
We tested on desktop, tablet and mobile, with the share of participants on each device matched to what Middlesex’s analytics showed. Tasks were device-agnostic, so we could compare performance across devices. Every participant completed core tasks, and each group also completed tasks specific to them, such as finding a course, comparing accommodation or searching for research.
A consultant observed each session from a separate observation room, so their presence didn’t affect behaviour, and every session was recorded for further analysis. Middlesex staff could watch live.
“Both the service and project execution were fantastic – we were consulted at every stage in a timely fashion, and couldn’t have felt more welcome. It’s been a fantastic experience.”
Maria Luisa Ross, Head of Programmes and Insight, Middlesex University
What we delivered
- Task efficiency scores and success rates for every task in the plan.
- Findings from free exploration of the site.
- Prioritised recommendations, ranked by urgency and likely impact.
- A System Usability Scale (SUS) benchmark to measure future changes against.
The outcome
Middlesex came away with a measure of how well its redesign worked for each audience on each device, and a SUS score to judge future changes against. With recommendations ranked by urgency and likely impact, it could plan its next round of improvements starting with what mattered most.
Frequently asked questions
Test with each of the audiences it serves, on the devices they use. Middlesex University’s testing covered four audiences on desktop, tablet and mobile.
A standard ten-question survey that gives a usability score out of 100. It’s useful as a benchmark to compare against after changes.
Yes, as well as before. Testing after launch shows how the new site works with real users on real devices, and gives you a benchmark to measure future changes against. Middlesex University’s testing produced both.
Because people behave differently when someone is watching over their shoulder. Observing remotely gives more natural results.
What this means
for your organisation
If you’ve just redesigned your website, or are about to, the surest way to know it works for every audience is to watch them use it. Analytics tell you what happened; testing tells you why.
If you serve several very different audiences, send us a few lines about them. We’ll tell you how we’d test your site with them.
Talk to one of our digital experts
Sophie Collins
UX Designer
Do you need help maintaining product quality and business alignment?
Passionate about all things User Experience and Agile, Sophie has worked across the entire software delivery lifecycle.
With a background in Product Ownership and Software Testing, Sophie approaches UX with both pragmatism and quality, focusing on producing results which have value for both businesses and users.
Or call us on 020 7439 1900
Related case studies
-
TBC Bank Usability Testing
Read more: TBC Bank Usability TestingRealistic kiosk testing that found problems before development began
-
Environment Agency
Read more: Environment AgencyFishing licence usability testing, rated full marks for client satisfaction
Have a project you’d like to discuss?
Give us a call on 020 7439 1900 or fill in the form and we will get back to you.