Creating a staff guide
Consolidating staff information from across the main University of Kent website to create a ‘Staff Guide’.
Discovery
The University of Kent website was devolved into an organisational layout, with each department having its own set of branded webpages. Whilst the branding on these pages was the same, the structure and tone was left up to individual departments to manage.
The challenge: Staff members at Kent regularly complained about: being unable to find relevant information where they expected it; and the main site search serving too much irrelevant information featuring the keyword(s) they used.
The solution: Create a one-stop shop for finding all staff information regardless of department of origin, organise it in a logical and task-oriented structure, and lock down the search to just this collection of pages.
Research
Competitor analysis: I searched through at least 30 different staff sites from higher education institutions to explore how they approached the problem.
Content audit: I conducted an extensive content audit across the whole of the University of Kent website, around 400,000 individual pages, cataloging and sorting staff related content as I went.
Surveys: Two online surveys were conducted, open to all staff members. Their aim was to collect as much information as possible on pain-points or successes when accessing staff information on the Kent website.
Focus groups/interviews: I organised and managed four focus groups, each with around 12 attendees, in three countries: England, France and Belgium. Similarly I also carried out 21 one-to-one interviews, spread out across six weeks, with a cross-section of staff members: part-time receptionists, academics, managers, administrators. The aim was to connect with the audience base and learn what was working and what wasn’t. This also provided an ideal setting for card sorting exercises and to try out some early prototypes.
Inspiration
During the discovery phase of the project I began to notice that universities with a more user-friendly approach had focused on a task-oriented structure. Similarly, many of the users I encountered during my focus groups cited Kent’s Student Guide, a site with a task-oriented structure, as something that should be emulated for the Staff Guide.
Experience and design
The primary purpose of users accessing the site is to search and find relevant information quickly. The secondary purpose is to access quick links to systems and services they need to perform a task.
User journeys and information architecture
I traced the user journey of a staff member’s time at Kent from accepting a job offer through to leaving Kent and used that as a way to identify when they might need specific information. The content was then easily categorised to fit alongside that journey, creating the initial information architecture for the Guide. Any assumptions made during this stage were tested and reviewed through focus groups and one-to-one interviews with staff from different backgrounds.
Sketches and wireframes
Simple sketches and wireframes were created using pen and paper to trace out the complete user journey from the search field to specific content in as few interactions as possible. This was tested with a random assortment of users and refined and helped inform the early designs.
Search optimisation
One of the most repeated pieces of feedback received during the discovery phase was the poor experience of the main Kent website search (at the time). Keywords would surface content from all sorts of sources with minimal filtering options or focus. To ensure the absolute best experience that could be delivered, I decided that the Staff Guide search should only surface content that was on the Guide(with an option to apply the search to the rest of the site).
Using the keywords analytics from the main site, I made sure that those same keywords were included on the relevant page and meta description. This meant that the content matched our user’s choice of language as closely as possible, leading to a more intuitive experience.
Final Results
