The benefits of segmentation in student recruitment


At SMRS it’s always been a mantra of ours that research and insight should inform everything we advise our clients to do.

Ten years ago this might have been using postcode mapping to understand where our clients’ recruitment hotspots were – or what areas a university wasn’t attracting students from. And the insight gained from this would inform a targeted and relevant out of home, press and direct marketing media strategy.

Then digital tracking and targeting was added to the media mix. Increasingly sophisticated, this enabled us to use the data available from our campaigns to inform future digital strategy and media buying. And both approaches are still relevant and used today.

Segmenting a competitive advantage

As Universities find themselves in an increasingly competitive market, the need to differentiate themselves and their particular offer is vital. And, to help them do this, they need to identify:

- what factors affect the likelihood of applicants enrolling
- and how different ‘segments’ of students are influenced by these different factors

A pioneering client

Every fresh-thinking initiative needs a client willing to be the first to try it. And, although we’ve since worked with a number of clients on student segmentation projects it was Birkbeck who, after seeing the value of mapping and mosaic profiling of their students, took the leap.

Birkbeck doesn’t have any trouble in differentiating itself. It’s an evening-only institution. So they wanted to find out:

- who was more likely to enrol
- how to find more of them
- identify areas where changes in process or behaviour of the institution would improve conversion and retention

As, if they could do this, then they’d have a great chance of growing student numbers in the future.

How segmentation works

We were given Birkbeck’s anonymised applicant data from the previous five years. This contained demographic information – postcode, country of residence, age and gender. It also provided behavioural information and included data about applicants education, careers, date of application, date of offer, number of course applied for, and previous study at Birkbeck, amongst others.

It was a vast amount of data. But with the help of our expert Data Analyst and state-of-the-art software we began to make sense of it.

What we discovered

The initial outputs were decision trees – which can be hugely complex themselves. From these we were able to identify key trends and patterns. We used these findings, together with an excel tool that allows the user to change any of the variables in the applicant data, to identify propensity to enrol. This then created pen portraits of the most likely segments to enrol – at PG and UG level and dependent upon the mode of study (FT or PT).

The Birkbeck benefit

The client saw the outcomes the project as “potentially ground-breaking for the sector” and it has led to a number of internal initiatives to improve conversion and retention.

In addition, we’ve adjusted and become more targeted with our marketing strategies for Birkbeck and the student recruitment numbers for 2017 indicate that this project was more than worth undertaking, with performance improvements across the board.


Monday, 19 June 2017

Share this article

Any questions? Any comments?

Your instant reactions to this article can be posted here. Use your own name or a nom de plume.

(Showing 1 - 3 of 3)

paul roper Date: Jun 19, 2017

Intrigued to know which DMP you used - are you able to say? thanks Paul

Jon Kirk Date: Jun 26, 2017

Hi Paul The segments were built outside of a dmp using first party data alone. Jon

paul roper Date: Jun 28, 2017

Thanks Jon really appreciated! Paul

Please log in to make a comment

Already registered?

Haven't registered?

Register for FREE - it only takes a couple of minutes

Not registered? Click here.

The benefits of segmentation in student recruitment