Finding Anglia Ruskin University’s true audience

Anglia Ruskin University has been educating students in a variety of subjects since 1992. They knew they were attracting students from across the UK but still felt that they were over-reliant on their East of England pool of prospects. To help expand their geographic footprint, ARU needed to understand exactly who they were attracting, from both within their region and outside.

Working together, we started with a detailed machine learning analysis, looking at application and enrolment data from the last 3 years. Through a combination of statistical modelling techniques we were able to establish specific and targetable audience segments based on their propensity to enrol at ARU. We used decision trees to help us identify the most important variables impacting on their likelihood of applying. Once the relationship between the different variables was finalised we could create statistically significant segments.

This analysis allowed us to group our audiences by a range of characteristics such as faculty of interest, campus application and so much more. With 56 individual segments, we were able to match these against available market data, so we could accurately understand ARU’s market share and the size of each opportunity audience that we could target.

Now we could craft engagement strategies, specifically for our different audience groups and objectives. We created campaigns from these strategies, to attract more people from across the UK. We constantly monitored campaign performance, so we could check that we were achieving the results that we had modelled.

Our work paid off and in partnership with ARU, the results we saw were impressive:

• We attracted 34% more UCAS applications by the January UCAS deadline in 2021 than in 2020.
• Each faculty saw an increase of at least 19% in applications.
• Diversity of applications grew, with 1,103 more London applications, and 702 more South-East applications.
• And prior to that, Clearing applications grew by 36% in 2020 compared to 2019.