I am not the only one who remembers the magical promise and allure of lead scoring. A mystical number created by some mathematical algorithm that we presumed would unlock the secrets to save our budgets, find the right students sooner, and help solve the enrollment challenges we were all facing.
Of course, lead scoring has failed to meet those lofty expectations.
The truth is that the expectations were unfairly high, and this potential did not actually exist.
Here’s why: lead scoring is, at its core, purely an assessment of profile characteristics. Individual student variability means you’ll never get the meaning out of a mix of “gender, grades, location” (and more) that we all thought could be possible.
Indeed, it’s helpful to know that students outside a 78-mile radius of your campus are very unlikely to enroll. But you need to find the ones who will.
What’s missing in lead scoring as a concept is the actual student.
Let me explain.
Of course, the profile factors considered are all related to the student, but none include the student’s voice or perspectives. To be most valuable, a data-science tool intended to identify the right applicants needs to include that.
To effectively direct attention and recruiting efforts, a scoring model must be responsive, quick, and accurately monitor student engagement. Frankly, there is no way to understand that before you begin communicating with a student.
Not to nerd out too much – but it’s not just the inclusion of student engagement metrics but a high level of data science work that needs to be leveraged to understand the relationship between variables. Knowing that a campus visit is meaningful is good, but it needs to be viewed in combination with all of the other variables available to you.
When lead scoring came onto the scene, we thought we had an ingenious answer to a long-standing problem. Snapshot scores based on profile factors will never do that for us – but that doesn’t mean that data science wasn’t the answer.