Higher Ed’s Costliest EdTech Mistake Is Looking Backward

Should higher ed leaders be relieved with the recent enrollment rebound, or is the hard part yet to rear its head?

In a recent Tech In EdTech conversation, I spoke with Phil Hill, educational technology consultant, industry analyst, and publisher of the On EdTech newsletter. As a long-term analyst in higher ed, Phil raised some pretty interesting points about the enrollment rebound. The biggest risk in edtech right now is not a single platform, company, or trend. It is the tendency to make decisions with yesterday’s assumptions. In this blog, we’ll cover what the enrollment rebound is really telling us, where higher ed still gets online strategy wrong, why vendor risk deserves more scrutiny, and how AI should push institutions back to the academic core.

Higher Ed Enrollment Rebounds Aren’t the Same as Recovery

Hill’s first point was the one most likely to be misread.

The long-term demographic pressure on higher education has not gone away. Call it the demographic cliff or, in Hill’s more measured phrasing, a demographic hill, but the larger trend is still real and still likely to shape the sector for years. Some enrollments have improved over the past two to three years, and community colleges have seen especially visible gains after absorbing some of the steepest losses.

That sounds encouraging. It should not be confused with stability.

As he noted, a major part of that rebound is tied to the rise in dual enrollment, with high school students taking college-credit courses in growing numbers. That matters because it is not simply more enrollment. It is a different enrollment mix, with different implications for revenue, program design, student support, and institutional planning.

That distinction matters more than the headline number. For example, a campus can point to growth and still be serving a very different market than it was a few years ago. That is not bad news by default. But it is a warning against reading better numbers as proof that old models still work.

The harder question for institutional leaders is not whether enrollments are up. It is what kind of growth they are seeing, what is driving it, and whether their academic and technology strategies match that reality.

Students’ Needs Should Drive Modality, Not the Other Way Around

One of Hill’s strongest points was that institutions need to stop treating modality like an identity and start treating it like a toolkit. Students are not loyal to a format for its own sake. They are trying to make education fit into real lives. For many working adults, flexibility is still the baseline requirement. But there’s an added expectation that flexibility should not come at the cost of connection.

Students want access and engagement. They want stronger faculty interaction, more meaningful peer connections, and a clearer sense that they belong to a program rather than simply passing through a platform.

The old version of online learning often assumed that convenience was enough. It is not. A course can be easy to access and still feel thin, isolating, and forgettable.

The better strategic question for higher ed leaders to ask is not whether a program is online or face-to-face, but whether the mix of delivery, support, pacing, and interaction fits the student it is meant to serve. Sometimes that means hybrid. Sometimes it means stronger synchronous elements inside a largely asynchronous experience. Sometimes it means redesigning support structures rather than changing the format itself.

Online Education Is Still More Local Than Many Institutions Admit

There is another myth that higher ed keeps dragging around: the idea that online learning erases geography.

For a small number of national brands, maybe. Western Governors University and Southern New Hampshire University are obvious examples. But Hill made the more important point: for most institutions, online demand is still local. Many online students choose schools close to where they live, often within commuting distance or at least within the same state.

It changes how institutions think about competition.

Too many schools still approach online learning as if it automatically turns them into national players. Usually, it does not. It expands the ways they can serve their region. This works as a strategic advantage if institutions are honest enough to use it.

An institution with pre-existing employer relationships, community trust, and real regional relevance should lean into those.

Phil also pushed the geography point further, and this is where the market gets more complicated. Geography entails both physical distance and policy. State authorization, reciprocity agreements, and the mechanics behind NC-SARA shape what is practical, what is costly, and where growth is realistic. There is no single online market. There are many, and state-level policy has more influence than many institutions account for.

That means expansion decisions need to be based on actual market conditions, including the regulatory ones.

The LMS Is Not Obsolete. It Is Just Unfashionably Essential

Every few years, the industry finds a fresh reason to predict the end of the LMS. This time, the argument is that generative AI and coding agents will make platforms easy to replicate and hard to defend.

Hill has heard versions of this story before, and his pushback is convincing. Institutions do not pay LMS vendors just for software code. They pay for the infrastructure that keeps learning running at scale: security, accessibility, reporting, student rostering, assignment workflows, consistency, and a reliable front door for faculty and students.

None of that stops mattering because software development is changing.

The more interesting issue is whether institutions are evaluating platforms in a way that makes any strategic sense. Hill argued that most leaders understand, at least broadly, that learning happens across an ecosystem of tools. Then procurement starts, and the conversation collapses into feature counting.

A platform should not be judged only by what it does today. It should be judged by how it fits into the broader ecosystem, how credible the roadmap is, and whether the company behind it is positioned to support where the institution needs to go next. A tidy checklist feels objective. It is often just a neat way to make
backward-looking decisions.

Vendor Stability Is Not a Side Issue

One of the most practical parts of my conversation with Hill had nothing to do with product features. It had to do with financial health.

Too many institutions treat vendor finance as somebody else’s problem. Hill’s point was straightforward: the edtech market is shaped by ownership structure, debt, and pressure to reach profitability, and those forces directly affect the institutions buying from it.

For example, Anthology’s bankruptcy and return to the Blackboard name are a reminder that the financial structure behind a company can reshape the customer experience. Coursera presents a different version of the same issue. As its strategy has shifted toward corporate training and internally developed AI-enabled content, institutions have every reason to ask whether the partner they signed up with is still aligned with their needs

That doesn’t mean private equity is automatically bad. Institutions need to understand the business realities of the companies they rely on. If a vendor is heavily leveraged, restructuring, changing priorities, or chasing a different market, that is part of the risk profile.

The question campus leaders should be asking is plain enough: where is this company headed, and do we still want to go there with them?

AI Should Push Institutions Back to the Academic Core

The easiest mistake higher ed can make with AI is to reduce the conversation to two bad options: panic or overreaction.

Hill was blunt about the first wave of institutional response. Too much energy went into cheating detection and not enough into the deeper academic issues that generative AI exposed. The next phase has been more practical, with platforms helping faculty and instructional designers create rubrics, quizzes, and other course materials faster.

Useful? Yes. Sufficient? Not even close.

The larger opportunity is not administrative convenience. It is an assessment. If AI has made many traditional assessments easier to game, the real problem is that too many assessments were already shallow.

That was true before 2022. AI just stripped away the illusion that institutions had more time to deal with it.

The smart investment is not “put AI everywhere,” and it is not “ban it from everything.” It is to rethink assessment with more urgency than most institutions have shown so far. More authentic evaluation. Better evidence of learning. Stronger alignment between what students are asked to do and what institutions say matters.

That work is harder than buying a feature. It is also far more important.

Higher ed does not need more trend chasing. It needs better judgment.

What stayed with me from this conversation with Phil Hill is that institutions are most likely to make bad edtech decisions when they confuse short-term movement for long-term direction. The campuses that navigate this moment well will be the ones that look harder at who today’s students are, how learning really works, which vendors are stable and aligned, and where academic design has to change before technology can help.

For leaders sorting through enrollment pressure, platform decisions, vendor risk, and AI priorities, the full Tech in EdTech episode with Phil Hill is worth your time. Listen to it or watch it for the full conversation.