A key part of managing content and getting people to find it is understanding your audience. This is the basis of our approach to content creation within Digital and Creative Media.

But not everyone in a large organisation thinks about users; often they have competing priorities internally that make it hard to do this.

As part of the New Web Estate project, we’re keen to make sure that any future way of working considers our audiences from the start.

Insights to inform strategy

We’ve been working with a content strategy consultancy to gather and generate audience insights.

These insights will inform our digital content strategy but also help to steer how people at Sussex think about users by giving them an audience insights database to use.

The database will support people who write content at short notice with tight deadlines to access recent research and write user-centred copy.

The six key audience groups we’ve focused on for this are:

• alumni
• civic community
• current students
• prospective students
• staff
• researchers.

Gathering existing insights

For the first phase, we’ve collected insights from across the university and analysed these with the agency to identify gaps in our knowledge using a four-layer model:

• access layer – the channels and platforms people use
• information layer – what our audiences want and need from us
• emotion layer – understanding what motivates users
• influence – what influences and distracts them.

We discovered that while we have good knowledge about the information that audiences need and the channels they use, we often lack understanding of user motivations and emotions.

Generating fresh insight

To fill these gaps, the agency conducted stakeholder interviews, surveys, user interviews and focus groups with each audience. These were added to our existing insights and uploaded to the new insights database.

The idea is that fresh insight can be added to the database over time and build on the audience knowledge we have.

Next

The next step is to finish a working prototype of the actual front-end of the insights database (powered by a large language model similar to ChatGPT), which can examine the questions we ask it and give us detailed answers based on the data it has.

This saves content editors from having to trawl through cloud storage folders looking for what they need, and we can continue to add to the base of information to build out the insights that the model can work from.

Sharing insights within the University

If you have audience insights (surveys, polls, focus group outputs and so on) that you’d like to add to the database, email dcm@sussex.ac.uk.