The Outsider Moment

​

Over the past 35 years, I've worked across almost a dozen  industries - from enterprise computing systems to major appliances to mobile technology. But when I entered higher education a decade ago to lead strategic transformations at colleges, I encountered the most complex challenge of my career: enrollment management.

 

What I found was an industry defined by urgency - but constrained by outdated tools. Counselors were overloaded. Decisions were driven by instinct. Data rarely translated into clear action. Teams were expected to do more - with less clarity, less time, and often, less support. 

​

The first time I sat in an enrollment strategy meeting, I felt like I'd stumbled into someone else's playbook. We were talking about enrollment growth, and when I asked what had contributed to recent declines, the answer was simple: "If you want more students, you need to give us more money." When I asked what if there wasn't more money, the conversation hit a wall.

​

That was the moment I realized this wasn't about budget. It was about assumptions. Assumptions so deeply baked into the system that no one questioned them anymore. A playbook that was dependent upon a set of market conditions that didn't exist any longer.

​

What We Built and Why

What makes our story different is that we didn't build enroll ml for the industry. We built it for our own teams.

We deployed it, proved it, and used it successfully for two years before ever offering it more broadly. That's why we so often hear, "This is the tool I've always wanted to build." Because it was built in the trenches - under the daily weight of the enrollment calendar. enroll ml wasn't designed in theory. It was forged in practice.

​

What we discovered changed how I thought about how enrollment management worked - and how it could work. What we thought were the big signals of enrollment - FAFSA submissions, campus visits, application timing, the aid award, the predictive score - were actually the surface-level representations of other, deeper sets of behaviors. The real signals of interest. Behaviors that we couldn't track before, that we never knew existed until we had the raw compute power of AI to find them. These deeper signals not only gave us more insight, they gave us more time to react, and recover.  To work on student strategy, not just tasks.

​

The System That Thinks

Two years into using our own platform, something emerged that we hadn't anticipated. We weren't just seeing patterns anymore - we were understanding them. The machine learning models had evolved to do more than flag which students were leaning in. They could explain why a student was hesitating, what was driving their decision state, and most importantly, what specific action would help them move forward.

​

I remember looking at a student who had all the traditional signals we'd been taught to trust - campus visit, early FAFSA, multiple portal logins. But our behavioral analysis was showing warning signs. The reasoning engine broke it down: her engagement had actually peaked three weeks earlier, then steadily declined. Her logins were shorter. Her page views more scattered. The pattern suggested drift, not momentum.

​

But we needed a system that didn't just diagnose the problem, but recommended the solution. Not a generic "reach out to this student," but specific, context-aware guidance: "This student's concerns appear to center on academic fit. Her behavior suggests uncertainty about program rigor. Recommend faculty connection in her major within 48 hours."  We quickly realized that the advancements in AI deep-reasoning could provide us those answers.

​

That's when we moved beyond analytics into something entirely different: true personalization at scale. This wasn't just pattern recognition anymore. It was decision intelligence.

 

Beyond Guesswork

Today, enroll ml supports dozens of institutions across the country. And while machine learning powers the platform and deep-reasoning AI individualizes strategy and creates context-aware messaging, it doesn't replace the humans behind the work - it empowers them with personalization at scale, instantly available at their fingertips.

​

We believe in the humanization of enrollment management, and enroll ml is built to provide the insights that build stronger connections between the institution and the student, when they matter most. 

​

The result?

A better experience for institutions. Smarter, more focused teams. And most importantly - better outcomes for the students we serve.

​

Because when you can see not just what students are doing, but understand why they're doing it and know exactly how to respond, enrollment management stops being guesswork. The signal becomes clear. And teams can finally do what they came here to do: guide students, not chase ghosts.