I say it all the time. You probably do, too. Good enrollment management practice starts with good data, and good analysis of the data. It’s hard to argue that point.

The word “starts” is critical in that sentence, however. I’m often reminded of this graphic, which is a good one for people just getting started in Enrollment Management and data analysis.

You can’t take a bowl of flour and eggs and tell people you’ve given them information cake, let alone insight. Full disclosure: I often struggle to help people see the most compelling and salient data points among the thousands that might pop up in my analysis; it’s harder than it sounds.

I always like to see new takes on data and analysis, so I was intrigued by the post below I saw on LinkedIn recently. I am not trying to be critical of the poster here, as I think this is a great starting point for discussion. The analysis is almost certainly based on aggregated data as reported in the Common Data Set or other data sets, so the ability to get more granular is by definition limited.

Take a moment to look it over and see what you think about it. Then I’ll tell you what I see.

The problem I see is that it measures headcount, broken out by income level. There is, of course, great value to measuring headcount changes, as it’s a reasonable proxy for revenue, and it usually translates into staffing needs. First-year enrollment in the state of Ohio (top) has dropped 12% over four years, and the small institution (which I presume in also in Ohio, but it’s not clear) has lost 41% of its headcount in the first-year class over that time.

So, what’s the problem? If you haven’t figured it out, here’s a hint: I frequently tell people I can get as many students enrolled in their program as they want if I can control pricing. Cut the price by 90%, and you can triple your numbers. Is that a good business transaction?

Again, the analysis above is limited by the detail available to the researcher, but the danger is the easy conclusion that can be drawn by doing this sort of analysis on your campus, when you have the data but don’t know how to look at it.

Those low-income students may be commuters who bring a full Pell and a large state grant to the table. Those high income families might be coming because of record discounts as a college pursues outcomes like raising SAT scores, or opening new markets among mobile (read: wealthier) students.

Ask your enrollment manager to do this analysis by looking at the net revenue of these first-year students, both in aggregate and in average. Then ask them to look at that cohort in year two, and three, and four to calculate an NPV. With a few extra steps, you might realize that the first, top-line chart was dead on, and raised the right issues. In that case, you feel more confident about diagnosing and addressing the problem.

But maybe you’ll find something different. And if so, you’ll feel even better.

Enrollment is complicated. Context is critical.