What Does An Accurate Predictive Model Do For Admissions?

If you could eliminate 30-60 percent of your marketing spend while generating the same number of applications, would you do it? Maybe a better question is — why wouldn’t you do it?

That’s what a robust, sophisticated and, most importantly, highly accurate predictive model can offer colleges and universities. It’s a tool that schools can utilize to prioritize their efforts confidently — to get more applicants for the same amount of money.

Let’s walk through a scenario of how that would work.

Most students who receive a postcard from a school throw it immediately in the trash. But what if you knew the ones who would be interested? Then you could save some serious money and only send the postcard to students who would actually read and consider it.

Fortunately, there’s a way to find those students. You need a predictive model that performs accurately in the real world.

The applicant models generated by Capture Higher Ed’s comprehensive predictive engine, Envision, fit that description, and we have the proof. For Fall 2017 predictions, students who we predicted to be in the top 10 percent most likely to apply did end up applying at a high rate. In fact, they applied 7 times more often than students with lower predictions.

Below is a chart of our applicant rates across clients compared to our predictions — 10’s are predicted as the students most likely to apply. 1’s are predicted as the students least likely to apply. You can see that our predictions lined up with what happened in the real world.

Accurate Predictive Model

This translates to 98 percent prediction accuracy.

But what does an accurate predictive model do for an admissions office? It allows them to prioritize their time, effort and money confidently. Let’s put that in concrete terms.

If you had a 100,000-student suspect pool, it would cost $100,000 to send them all a postcard at $1 apiece for printing and mailing. But what if you knew the best students to send it to, the ones who would respond to your marketing? Then you could save some serious money on your direct mail budget.

An Envision applicant model can tell you that.

If you sent the postcard to only half of the pool, you would save $50,000 and still get 95 percent of the total response you would have gotten otherwise. Or if you’re even more frugal, you could send the postcard to just the top 10 percent, save $90,000 and still get almost half the response you would have otherwise.

Can you imagine what you could do with $90,000 more in your budget? Instead of spending it on a postcard that those extra students would throw in the trash, you could:

  • Buy more search names to communicate with through a low-cost medium like email;
  • Put it toward advertising on Facebook;
  • Send the very best students a glossy viewbook instead of just a postcard;
  • Give 18 students $5,000 more in scholarships. (Now there’s something that’s going to impact your class way more than a postcard!)

You can use an accurate predictive model to prioritize your money. You can also use it to prioritize your time. Our partner schools use their Envision models to decide which students they should target with a phone call, which ones they should write a hand-written note to, and which students to seek out on a high school visit.

This all translates to less waste, with the end result being more applicants for the same amount of time, effort and money.

By John Foster, Senior Data Analyst, Capture Higher Ed