The Future of Admissions Is The Middle, Not The Funnel

Most colleges have built a solid system at the top of the enrollment funnel. Marketing automation captures interest, digital campaigns drive inquiries, and CRMs fill up with names. At the bottom of the funnel, experienced counselors do what they’ve always done: build relationships, nudge undecided students, and bring in the class.

But between the two, something breaks.

This is the part of the funnel where students don’t always follow a clear pattern. Some apply early but go quiet. Others open every email but never schedule a call. A student may deposit and then disengage completely. These aren’t outliers—they’re normal behaviors. And this is where most teams struggle.

The middle of the funnel is where decisions get delayed, momentum stalls, and melt risk begins to grow. And it’s also where AI, when applied with focus, is starting to make a measurable difference.

Edgewood College saw this firsthand. Their team was strong, but overstretched. Applications were growing, but so was counselor fatigue. When they implemented machine learning specifically to manage mid-funnel engagement, the results were immediate. They increased enrollment by 25 percent and improved counselor productivity by 30 percent. Not because they worked harder—but because they worked on the right students.

That shift didn’t come from predicting who would enroll. It came from identifying who needed attention each day.

At its core, managing the middle of the funnel well means knowing who is moving forward, who is stuck, and who is slipping away. Traditional tools aren’t designed to answer those questions. CRMs collect data, but they don’t interpret it. Milestone-based reports are static. Counselors often rely on instinct or whatever list they exported last. And by the time that list is reviewed, the student behavior may have already changed.

What worked at Edgewood was a daily system that scored students based on behavior and timing. It flagged when a student’s actions were out of sync with their enrollment stage. It tracked momentum, not just milestones. And it freed up counselor time by removing the need to search for signals—they were already there, organized and ranked, every morning.

This is the practical role of AI in admissions—not automating decisions, but structuring counselor focus.

The idea isn’t that the middle of the funnel has been ignored. It’s that it has been underserved. Many of the tactics that institutions use—email follow-ups, virtual events, nudging campaigns—depend on someone deciding who should receive them. When that decision is based on intuition or spreadsheet sorting, good students get missed. When it’s based on time-sensitive behavioral signals, teams can reach students while their interest is still active.

And that’s the point. The middle of the funnel is defined by timing. Interest can change in a matter of days. Drift isn’t always visible in the CRM. And melt doesn’t start in August—it often begins in April, when a student starts responding more slowly or disengaging entirely.

This isn’t a theoretical concern. Roosevelt University used the same approach to manage summer melt. By detecting which depositors were no longer engaging, they focused outreach on the students who were actually reconsidering their plans. That change helped them keep more of the class they had already recruited.

If the top of the funnel is about reaching students, and the bottom is about confirming their decisions, then the middle is where most of the real enrollment work happens. It’s where teams win or lose yield. It’s where missed opportunities accumulate. And it’s where many counselors spend the most time—without the guidance they need to make that time count.

Machine learning isn’t valuable because it sounds advanced. It’s valuable when it makes the middle of the funnel manageable. That’s the work that is already delivering results. That’s where the next decade of enrollment strategy is being built.

Join us on August 26 to hear how leading institutions are using AI to manage the middle, reduce melt, and increase counselor capacity: crowdcast.io/c/vpjohn