Enrollment Strategy
By Angela Brown
You’ve probably noticed that every enrollment technology vendor is "AI-powered" now.
It started slowly, and then it was everywhere. Press releases, pitch decks, conference booth signage. "Intelligent orchestration." "Predictive engagement." The jargon multiplied faster than wet Gremlins.
If you’re confused, don’t worry, because most of those phrases don't mean anything specific. And in higher ed, where enrollment decisions are high-stakes and budgets are far from unlimited, "it sounds cool" isn’t a procurement strategy.
This post will give you some guidance on how to separate the nonsense from what really works, particularly as you evaluate enrollment technology going into the next planning cycle.
What "AI-Driven Orchestration" Actually Does (or Doesn't)
Let's start with a phrase you've probably seen: AI-driven orchestration.
Ohhhh. It sounds sophisticated right? It implies something intelligent is managing the cadence of your student communications, routing the right message to the right student at the right time. And that would be pretty valuable.
But what, exactly, is the AI deciding? And how?
If you get a vague answer — "the system learns over time" or "we use machine learning to optimize campaigns" — that's a red flag. It means the AI is doing something, somewhere, but the vendor either can't or won't explain what.
Orchestration without transparency isn't intelligence, it's a black box. And black boxes are fine until enrollment targets aren't met and you need to understand why.
The Difference Between AI Vocabulary and AI Functionality
Here's a useful mental model for higher ed technology evaluation.
Think of it as a spectrum. On one end you have AI vocabulary. On the other, AI functionality.
AI vocabulary is the language vendors use in their marketing. It's aspirational, sometimes accurate, almost always optimistic. "AI-powered." "Smart segmentation." "Dynamic personalization." These terms are technically defensible in almost any context because they're never precisely defined.
AI functionality is what the system does. Specifically, demonstrably, and repeatably. It either makes a decision or it doesn't. It either shows its work or it doesn't. It either improves based on your data or it doesn't.
The evaluation question should be: which side of that spectrum is this vendor on?
Most are somewhere in the middle, which is fine. But a vendor whose AI messaging lives entirely on the vocabulary side, with no concrete explanation of the underlying mechanics, should raise a flag during your enrollment technology evaluation process.
Why the Knowledge Base Question Is the One Nobody's Asking
Now for the question that doesn’t get enough attention in demos. When an AI system communicates with prospective students, where is it pulling its information from? Who updates it? How does it stay accurate across program changes, scholarship updates, and policy shifts that happen mid-cycle?
All of that ladders up to a knowledge base, and it matters significantly in an AI-first enrollment environment.
A lot of enrollment platforms were built as point solutions: a CRM here, a chatbot there, a texting tool bolted on. And everyone has gone all in on agents. Each one has its own data layer, its own inputs, and its own version of the truth about what your institution offers. When AI runs across those disconnected systems, it's working from fragmented, often stale information.
The result is a student who asks a question about financial aid, gets one answer from the AI chat tool, a different answer from the email campaign, and a third answer when they finally call admissions. That's not AI doing its job. It’s only making your problem worse and harder to trace.
A unified vertical stack, where the knowledge base, the engagement layer, and the AI are all operating from the same source of truth is the only architecture that scales in an AI-driven enrollment operation. Without it, you're managing inconsistencies instead of a system.
Questions to Ask in Every Enrollment Technology Demo
This is the part you can take into your next vendor call.
On AI transparency:
- What specific decisions is the AI making in your platform?
- Can you show me a real example, not a hypothetical, of the AI changing a communication based on student behavior?
- If a campaign underperforms, how do I find out why?
On the knowledge base:
- Where does the information your AI uses come from?
- When program details change, what's the process for updating that data across the platform?
- If your AI is integrated with other tools in our stack, how do you prevent information conflicts?
On outcomes:
- What does your AI optimize for — opens, replies, enrollments, something else?
- Can you show enrollment or yield outcomes attributable to the AI specifically, not the platform generally?
If a vendor gets defensive or deflects on any of these, that tells you something. The best enrollment technology partners answer these questions quickly, specifically, and with data.
Transparent AI vs. Black-Box AI: What It Looks Like in Practice
There are two fundamentally different approaches to AI in enrollment technology.
The first is opaque by design. The AI runs, the outputs appear, and the vendor tells you to trust the model. You can see what the system sent and when, but you can't always see why, what it predicted, or how confident it was. Reporting exists, but explaining the AI's reasoning requires a support ticket or a quarterly business review.
The second approach puts the reasoning in front of you. Before a campaign goes out, you can see what the AI predicts will happen: which students are likely to engage, what behavior it expects, and where it thinks yield risk is highest. After the campaign runs, you can compare prediction to reality. The model is accountable because you can see it.
The second approach is harder to build. It requires the platform to commit to a prediction, which means it can be wrong, and that means the vendor has to be confident enough in their AI to let you evaluate it.
That's the difference between AI as a feature claim and AI as a working system.
How to Evaluate Strategic Enrollment Technology Without Getting Played
A few principles worth keeping in mind as you go into any higher ed technology evaluation this cycle:
Vague AI claims scale with budget. The more expensive the platform, the more elaborate the AI vocabulary tends to be. That's not a coincidence. Big contract renewals don't survive "we're not sure what the AI is actually doing."
Ask for the methodology, not the case study. Case studies show you best-case outcomes under ideal conditions. Methodology shows you how the system works under normal conditions, including yours.
Integration promises are not integration realities. A vendor saying their platform "integrates with your existing stack" is not the same as having a unified architecture. Ask specifically how data flows between systems and who owns the knowledge base that feeds the AI.
Simulation beats promises. If a vendor can show you how their AI would respond to your student population before you sign, that's evidence. If they can't, you're buying a hypothesis.
Strategic enrollment management is too high-stakes for hypothesis-based procurement. The demographic cliff is happening, the competition for students is real, and the difference between vendors who have genuinely functional AI and vendors who have impressive AI language is bigger than most demos reveal.
Ask the hard questions, push past the vocabulary, and demand to see the work.
________________
As Halda’s Director of Marketing, Angela Brown brings more than 15 years of experience leading marketing and content teams in education and B2B SaaS. When she isn’t at her computer, you can find her reading, watching a true crime documentary, or driving her son to basketball practice.
