For months, my team worked on a project that we called program page optimization—and when I say months, I mean months. We rewrote program pages. We changed their architecture. We reconsidered images, content hierarchy, calls to action and the overall purpose of the pages. It was a heavy lift for a small team, but it was necessary.
Over time, some of our academic program pages had become something closer to storage space. They held information important to faculty, passion projects, historical content and whatever else had accumulated over the years. One program page could look entirely different from the next. Somewhere along the way, the person we most needed the page to serve—the prospective student—could get lost. So, we changed that. We built consistency. We thought about search. We thought about the questions students were asking and how someone moved from discovering a program to understanding it and, ultimately, taking the next step. We did the SEO work, then came the next conversation.
“You're doing a lot right,” our partners told us. “But prospective students are increasingly using large language models to research colleges.” Okay, we knew that. “And they may never actually visit your website.” I'm sorry. What?
After months of optimizing pages so students could find our website, we now had to consider a world in which they might form opinions about us, compare us with competitors and even eliminate us from consideration without ever getting there. Welcome to enrollment marketing in the age of AI.
The Search Is Happening Before the Click
For years, much of digital enrollment marketing has been built around a familiar journey. A prospective student searches, sees your institution, clicks a link and arrives on your website. From there, we work very hard to move that visitor toward an action. Increasingly, however, the search itself can be the experience.
A prospective student might ask an AI tool, “What college offers the best price for an online undergraduate degree?” Then:
- “Which ones are accredited?”
- “Which offer financial aid?”
- “Which can I finish while working full time?”
- “Which one is best for someone like me?”
The student hasn't filled out a form. They haven't become an inquiry. They may not have visited a single .edu website, yet they are already evaluating institutions.
Research suggests this behavior is moving quickly. EAB’s 2026 survey of more than 5,000 high school students found that 46% were using AI tools during the college search, up from 26% in spring 2025. More strikingly, 18% said they had removed a college from consideration based on information that surfaced through AI-generated search results.
The shift extends beyond traditional undergraduate recruitment, but AI isn't replacing other search behaviors. Research from UPCEA and Search Influence found that half of prospective adult learners surveyed use AI tools at least weekly. At the same time, 84% still use traditional search engines and 63% visit university websites while exploring their options. That distinction matters.
SEO isn't dead. The website isn't dead. Students are simply adding new ways to discover, compare and evaluate us. The question is no longer simply, “Can students find us?” It is also, “What will they learn about us before they ever reach us?”
Enter the Wizards Behind the Wi-Fi
I am not going to pretend that every enrollment leader needs to understand exactly what is happening behind every AI-generated response. There are bots and crawlers and algorithms and, as far as I'm concerned, a few machine magicians doing mysterious things behind the Wi-Fi. What I do need to understand is what this new search environment requires from us.
Consider something as fundamental as price. A student might ask, “What are the most affordable online bachelor's degrees in Texas?” Can an AI tool easily determine what your program costs? Not what your tuition page implies. Not what someone can calculate after opening three pages, finding a PDF and doing some multiplication. What does it cost? The same applies to financial aid, programs, outcomes, admissions requirements, modality and time to completion. If we make those answers difficult for humans to find and understand, we should not be surprised when AI struggles to interpret them accurately. Suddenly, content architecture isn't simply a website decision. It is part of whether an institution can be accurately understood in an AI-mediated search environment.
Breaking Some Old Habits
We didn't throw away everything we had learned about SEO. We just discovered we weren't finished. Job security, I suppose. Our team began thinking differently not only about program pages, but about all of our web content. We are still working on it because search behavior is evolving while we are responding to it. Some of the shifts sound subtle, but they represent a significant change in mindset.
Old habit: Optimize for keywords.
New habit: Optimize for questions and conversations.
Keywords still matter, but now we are thinking more intentionally about the questions prospective students ask and whether our content provides clear, direct answers.
Old habit: Get them to the website.
New habit: Make sure the institution is accurately understood before they reach the website.
Traffic still matters. Conversion still matters. The website absolutely still matters. In fact, the UPCEA research found that 77% of prospective adult learners surveyed considered university websites the most reliable source of information, but discovery increasingly happens in places we do not own.
Pew Research Center found that when Google users encountered an AI-generated summary, they clicked a traditional search result in only 8% of visits, compared with 15% when no AI summary appeared. Just 1% clicked a source link within the AI summary itself. The zero-click search isn't theoretical. The behavior is already measurable.
Old habit: We control our story because we control our website.
New habit: Our entire digital footprint helps AI assemble our story.
Consistency matters. Third-party information matters. Clear answers matter. Authority matters. And increasingly, generative engine optimization and answer engine optimization matter alongside traditional SEO.
GEO Is Enrollment Strategy
It would be easy to categorize GEO as another digital marketing tactic. Give it to the web team, add it to the SEO checklist, learn a few new acronyms and keep moving. That's not enough.
If a prospective student can ask an AI tool to compare institutions based on price, programs, financial aid, outcomes or fit, then use that answer to decide which institutions deserve further consideration, GEO is affecting enrollment before the traditional funnel even begins. Before the inquiry is submitted. Before the campus visit is scheduled. Before the application is started. Maybe even before the click. That makes GEO an enrollment strategy.
A student can discover an institution, compare it with competitors, ask follow-up questions and narrow a list without ever identifying themselves to us. If the information AI finds about our institution is unclear, inconsistent or incomplete, we may never know that student considered us at all. We simply don't make the list.
None of this means that the SEO work was meaningless—quite the opposite. Good SEO forced us to become clearer, more organized and more student-centered. GEO and AEO are pushing that work further.
Higher education has a remarkable capacity for making simple things complicated. We create language students don't use, organize websites according to our organizational charts and occasionally require prospective students to become amateur detectives just to determine what something costs. AI search is exposing the weaknesses in that approach. For enrollment and marketing leaders, GEO is not optional.
That doesn't mean chasing every new acronym or rebuilding your digital strategy every six months. It means recognizing that prospective students’ search behavior has already changed, and our practices must change with it.
I started this journey thinking we were optimizing program pages. What we were really doing was learning how to make our institution easier to understand. SEO taught us to think about how people find us. GEO and AEO are forcing us to think about whether they can understand us, sometimes before they ever arrive. So, yes, after months of program page optimization, I discovered there was more work to do. Of course there was. But if the result is clearer information, better answers and fewer prospective students trying to decode higher education jargon while a machine magician somewhere behind the Wi-Fi does the same thing, I'll take it.