Navigating through MDU (Mälardalen University)
Enabling students to find vital information within their own information-packed university website by navigation system and information re-design.
Guerilla Testing • Agile UX • UX Research • UX/UI Design • Industry-based project
My role & responsibilities
Agile: Scrum master, Planning, Creating epics and user stories
UX Research: Quantitative Data Research, Heuristic Evaluation, Guerilla Testing, Usability Testing, Affinity Map, Persona, Card sorting
UX/UI: Wireframes, Low-fi to Hi-fi prototyping
I led the methods in bold.
Project context
Client: Mälardalen University + HiQ (consultants)
Sep - Oct 2022, 4 one-week sprints
Agile industry-based project
Course: Serviceable Design @MDU
Team: Albin Pålsson, Jonna Edlund, Karolina Meijer, Rana Alostaz, Stephanie El-Hage & Sabina Lööf
Tools
Pen and paper
Figma + FigJam
Video recording
Swedish fika
Problem
Users fail to find key university info on MDU website, even on Google. Established students struggles more compared to other groups
Users can’t find how where to get mental support from the school and course information. Current students have a more difficult time to navigate through the MDU website than potential students. How might we ensure students to find their wanted final destination?
Background + Goal
MDU’s website fails 9 out of 12 benchmarks compared to similar higher ed. websites in an already critical media wave.
Each year, +20.000 students attend MDU programmes and courses, accounting for 7% of all student attendees in Sweden, who get in contact with the MDU website.
Spring of ‘22: MDU’s website participate in a website survey, getting benchmarked with 9 other higher education institutions in Sweden in the categories: structure, function, design and general impressions. The website ended up failing the benchmark in 9 out of 12 categories. MDU and the website product owner turned to us in hope of figuring out how to raise the benchmarks.
In addition, as MDU became Sweden’s newest university in 2022, they also had the spotlight on them to prove themselves in a critical media environment, questioning MDU becoming a university.
Goal: to research the benchmarks and target group(s) of our choosing, hopefully improving results.
Challenges
Usage of third party solutions, e.g. internal search engine, couldn’t be tailored to user needs
Working in agile industry-based project, we bumped into several challenges after discussing with our product owner, which we had to overcome and work around allowing us to present feasible and tangible solutions. These were the most important challenges we identified:
Our approach
With a 0.9% survey response frequency, I called for research-heavy pre- & first sprint with triangulation to enhance credibility of findings
At the project handover, our product owner gave us quantitative data to analyse:
As there was only a 0.9% answer frequency on the survey I persuaded our team to use triangulation to enhance the credibility of our findings and analysis. We used:
Quantitative data: Benchmarking + Google Analytics
Expert check: Heuristic evaluation, Competitive analysis (of universities ranking at the top of the same survey)
Qualitative data: Usability test with current website
Our sprints went from research-heavy to iterating from the insights we got from tests.
First findings + Paper prototype
Guerilla tests showed that current students anticipated a menu, need for categorisation. Expresses relief of not having to forever scroll
We decided to explore the benchmark categories: navigation and well-structured, as patterns in our research findings showed it to be a pivotal area to explore. We narrowed our target group to current students and started to explore the Student web, specifically aimed at students.
Our first guerilla test included giving the tester assignments based on the user stories from the survey, such as finding Student Health. Examples:
There’s been a lot in school lately, you need someone to talk to and suspect there’s help to be found in school. Where do you search?
You have a class at the other campus on Wednesday 10:15 and need a way to get there, is there any support our school can give for you to get there?
Information Design
Letting our users decide word associations by categorising themes with card sorting. Hope to use the language of our users
As we narrowed our scope we decided to create a menu for the Student web, it was very important to get the categorisation of the content correct as we would group up all main content from the Student Web into a menu. We decided to do a card sorting with 7 students, letting them group up subjects and then name their groupings.
UI Design
Users find their searches with great success when using our menu🥳However, they need support to actually find our menu…
We applied our new groupings for a whole-screen menu prototype with the attempt of shutting out surrounding information. The learning curve of finding items in the menu was easy, however the users couldn’t find the menu in the first place.
I argued to my group that they won’t find the menu because the student web look too similar to the home-page and as they didn’t use a menu on the home-page, they won’t suspect using one on the student web.
Overlay menu in first hi-fi iteration
Our first hifi-prototype was in grayscale as we wanted our users to stay focused on function, not form
We used our new card-sorting groupings to great effect!
However, the users had difficulties finding the hamburger menu. Even when adding two different icons to really see if it was doomed
Solution
80% of testers wanted to use our solution, everyone thought it was easy to use with no need to scroll
Before: Students using return, a lot, when trying to find correct link.
Our triangulation showed a lot of scrolling needed to be used, something our users were tired of doing
Scrolling through long articles to find the correct section created cognitive fatigue
Navigation links wasn’t using the users language. Words such as “VFU” and “Aj, oj halloj!” left the users confused
After: Drop-down mega-menu let users find the information, where they expect it!
We needed a new menu that was in-your-face, and so our drop-down menu was born!
We created flexibility by adding ambivalent words in more than one category as our tests showed that one-size doesn’t fit all
The sticky side menu made the user feel they saved time and effort
Impact
“During next spring we’re starting a sprint, grounded in on your case. The work you’ve done has been a crucial part of that process!” - Product owner @MDU
Our project was so successful with both our users and clients that we got to present it after the course to the MDU Programming, Communications and Marketing team.
For a full statement: click the accordion below.
-
“I gave your group a relative open case and I’m very satisfied with the presentation and results you arrived at. As a client I experienced that you took the problem of the case, looked at it from multiple angles and scenarios to come up with a focus. Great conclusions and sketches to make something concrete of. It’s easy to work with you. You’re great at receiving feedback and guidience to further develop your sketches/product.
During next spring we’ll start a mega-menu sprint, founded on your findings and case. The work you’ve done has been a vital part of that process!…
…I’ve heard from the consultants at HiQ that you were fun to work together with in this case. You were professional, perceptive and took their recommendations to heart in a satisfactory fashion through development.” - Product owner @MDU, Dec 2022
Reflections
Guerilla testing taught me a lot. How to prepare for tests and being cautious not to lead the user
What I learned:
Always prepare usability tests with a test-run. Started my first tests without doing much preparation and found myself asking the user leading questions. During the last sprints I prepared and felt more confidence in being there for my user and being observant, picking up small interactions and non-verbal communications.
Call early for delegation of work to increase efficiency. At the start the group was uncomfortable and worked on a lot of things together as 6 people. However, this proved to be inefficient when we were iterating designs and doing other prep-work at the same time. Efficiency increased by calling out a need of splitting tasks early.
Swedish fika + Guerilla test = true! Who could’ve thought people liked cookies and coffee in the morning? In total we were able to get 44 unique guerilla testers for our project!