Educator Analytics: Can Schools Anticipate When A Student Will Enroll?

One common issue we hear from many of our partner educators is that they have difficulty anticipating when student enrollment will occur. Some classes fill weeks in advance, while others seem to have waves of student interest at the very last minute. All of this makes effective planning and forecasting more than a little tricky and stressful, to say the least.

Understanding when students will sign up for classes is hugely important. It has major implications for a school’s bottom line and stretches across departments,  from marketing (how can we market when students are in the right phase of their evaluation process?) to programming (when is it most appropriate to cancel a class, or consider adding a new section?).

To  help shed some light on these questions, we turned to the data and analyzed over 30,000 enrollments made via CourseHorse over the past year.

Specifically, we were trying to find out how far in advance students sign up for a class before its posted start date and what this may indicate for how a school manages its programs. 

Time Between Purchase Date and Class Start

The first thing we found was that the average student signs up for a class 17 days before its start date (see chart below). That said, enrollments are highly concentrated within 1 week of class, and there exists a step-wise decline for every subsequent week. Only 5% of enrollments came more than 8 weeks before the schedule start of class.  

time_between

Next, we went a level deeper and dug into the data on a daily basis in order to see if there were any insights that could be gleaned from analyzing the immediate 21 days before the start of class (see below). 

We were surprised to find such high levels of last-minute enrollments — nearly 8% of students sign ups were within 24 hours of class start, and 3% on the same day! In total, almost a quarter of students (23%) enrolled within 3 of the start of class. A substantial majority (59%) registered within 14 days of the start of class.

Days_between

Price Matters

As with life, generalizations from data can be dangerous.

So as a final step, we decided to look at how purchase behavior changes based on the price of a class. We plotted the time to class (Days from Purchase to Class Start) against the price of the class (Class Price).  The results provided strong support for our hypothesis that the price of class is a significant factor for anticipating when a student will sign up for a class.

As you can see in the chart below, as the price of a class increases, so does the length of time that students will sign up before the start date. While classes under $50 see an average of 11 days, those over $1000 are booked a full 4 weeks in advance. 

Classes_by_Price

What’s this mean to me?

As you consider these findings, a few important questions come up for educators. Each has important implications that may impact your school’s ability to maximize student enrollment and revenue: 

  • When do you currently cancel classes? Do you account for the fact that 23% of enrollments come within 3 days of class? Is this policy differentiated by price of class?
  • How do you predict when you may fill a class and have demand to offer an additional session? Are you leaving money on the table by not providing flexibility to create additional sessions of classes that demonstrate early popularity?
  • Does your marketing plan account for different enrollment behavior between higher and lower priced courses?

Does this reflect the experience of your school? As always, your thoughts and feedback is welcome and encouraged. Please feel free to leave a comment below, or reach out to the author at [email protected] 

About the Author: Evan Fain is VP of Education for CourseHorse. In this role, he leads the team that partners directly with over 1,250 leading educational institutions around the country, including colleges, universities, museums, non-profits and local learning centers. He’s also a big-time data nerd. You can reach him at [email protected]