AI Will Destroy Your University Website Traffic. AI Is Also Your Greatest Opportunity.

The Problem Is Already Here — It Just Hasn't Hit Your Dashboard Yet

If you are responsible for enrollment, digital strategy, or institutional communications at a university or college right now, you are managing a compound pressure that did not exist three years ago.

Enrollment targets are not getting softer. The prospective student population you are competing for is smaller in many markets, more internationally distributed, and more demanding of the experience of even researching your institution — not just attending it. Your peers are investing in digital. The board expects results. And the AI mandate has arrived at your institution in whatever form it arrived: a working group, a task force, a vice-president with a remit and no roadmap.

In the middle of all of this, something structural is happening to how prospective students find, evaluate, and shortlist universities — and most institutions have not fully reckoned with it yet.

Search traffic to institutional websites is declining. Not because your SEO is broken. Because the research process that used to drive prospective students to your program pages, your campus life sections, your tuition and funding pages — that process is migrating into AI. Students are asking an AI assistant which programs are worth applying to. They are asking which universities have strong research reputations in a specific field. They are asking whether your location, your culture, and your outcomes match what they are looking for in a career. And in many of those conversations, they are getting synthesized answers — and not clicking through to your website at all.

This is the threat. And it is real.

But the same structural shift that is hollowing out organic search traffic for institutions that ignore it is creating an advantage of unusual size for the institutions that understand it. The opportunity is not incremental — it is a redefinition of how an institution can reach a prospective student at the moment of highest intent, with the right information, before competitors have figured out how to show up in that conversation.

The question is not whether to engage with this shift. It is whether to engage early, with evidence, or late, with urgency.

How Prospective Student Search Behavior Is Changing

The model that most university digital strategies were built on — produce authoritative content, rank for relevant search terms, convert organic visitors through program pages and CTAs — still works. It just works less every quarter than it did the quarter before.

Google's AI Overviews synthesize answers directly in the search results, reducing click-through rates for queries where AI can generate a sufficient answer without sending the user to a source page. Research tools built on language models — Perplexity, ChatGPT's search mode, Claude — answer open-ended research questions with synthesized responses that pull from multiple sources, surface what they consider authoritative, and frequently do not require the user to visit any institutional page at all.

A prospective undergraduate asking "which Canadian universities have strong environmental science programs?" no longer necessarily lands on your program page. They may get a synthesized list, with brief characterizations of each program, before ever seeing your URL. Whether you appear in that answer — and how you are characterized — is no longer determined solely by your search rankings. It is determined by what the AI has absorbed from across the web about your institution, your programs, and your reputation.

This is not a technical problem you can solve with a prompt. It is a content and experience problem. The AI is summarizing what is already out there about you. If what is out there is thin, outdated, or fragmented — if your program pages were written for search bots rather than curious students — the AI will produce a thin, outdated, fragmented characterization. And the student will move on to the institution whose digital presence gave the AI better material to work with.

Devi Parikh, CEO of Yutori and former AI research leader at Meta, argued on the Product Impact Podcast that AI agents are already changing the fundamental structure of search — agents that monitor, negotiate, and navigate on behalf of users, making decisions about what information to surface and what to filter out before the user ever has to make a choice. For prospective students, the agent is becoming the first advisor in the enrollment journey. Whether your institution shows up as a credible recommendation or gets filtered out depends on the quality and richness of the digital signals you produce.

The Opportunity — Which Most Institutions Are Not Positioned to Capture

Every structural disruption produces winners and losers, and the distribution of outcomes in higher education's AI transition will be highly uneven — not because some institutions have better AI strategies, but because some institutions understand their prospective students' decision journey well enough to respond to this shift, and most do not.

The institutions that will win in an AI-mediated recruitment environment are not the ones with the biggest content teams or the most sophisticated technical infrastructure. They are the ones that can answer, with evidence from actual prospective student behavior, the following questions:

What questions are prospective students asking that your current digital presence does not answer? Most institutional websites are organized around the institution's internal logic — schools, faculties, programs, administrative units. Prospective students do not think in those categories. They think about careers, communities, costs, options, and outcomes. If your digital presence is organized around your org chart rather than your applicant's decision journey, AI models will surface your content in the wrong context — or not at all.

Where in the decision journey does your institution need to be present — and where is it currently absent? A prospective student researching graduate programs in clinical psychology does not make a decision in a single search session. They research over months, across multiple queries, comparing programs, funding packages, faculty, and outcomes. The institutions that show up at every moment of that journey — not just in the program directory search — are the ones that enter the applicant's shortlist. Mapping that journey, in detail, from actual student behavior, is the research work that tells you where your gaps are.

What is the experience like when a student does land on your site? The AI may send them there. The question is what happens when they arrive. Most university websites were not designed for the expectations of a student who has already received a synthesized overview from an AI assistant and is arriving with a specific question that needs a specific answer. If the experience is a navigation maze built around the institution's internal taxonomy, you lose the student at the moment they were most ready to engage.

Nicholas Holland, SVP of Product and Head of AI at HubSpot, made the stakes of this shift explicit on the Product Impact Podcast — as LLMs start deciding who shows up, the organizations that have not restructured their content and acquisition strategy around how AI surfaces information will find themselves increasingly invisible, regardless of how strong their underlying offering is. The signal problem is not about quality. It is about whether the right signals are in the right places for an AI model to find them and characterize them accurately.

What This Requires — and Why Most Institutions Are Not Ready

Responding to this shift requires a different kind of work than most university digital programs have invested in.

It is not primarily a technology project. Buying a new CMS or a personalization platform will not solve a problem that is fundamentally about how well your institution understands the student decision journey and how well your digital presence serves students at each moment in that journey.

It is not primarily a content project. Producing more program pages will not solve a problem rooted in the fact that existing content is organized for the institution, not the student.

It is a research project first. The starting point is a rigorous, behaviorally grounded understanding of how prospective students — across segments, across program types, across domestic and international profiles — actually research, evaluate, and decide. That understanding produces a decision journey map specific to your institution and your applicant population. The map reveals where your digital presence is serving students well, where it is absent, and where it is actively working against the student's research process.

When PH1 conducted research and service design work for Simon Fraser University, the process involved 45+ usability interviews with researchers, faculty, and students — generating a deep behavioral picture of the jobs people were actually trying to perform, and a prioritized sitemap built on customer jobs rather than institutional structure. The output was not a new website. It was an evidence-based direction for what the website needed to do, for which audiences, in which moments, to move the outcomes that mattered.

That kind of research is not a luxury for institutions under enrollment pressure. It is the prerequisite for every subsequent investment in digital — because without it, the investment is aimed at the institution's assumptions about student behavior, not at student behavior itself.

The Digital Acceleration Pillars — Applied to Higher Education

PH1 structures every university digital transformation engagement around four shifts — the Digital Acceleration Pillars — that distinguish institutions compounding their digital advantage from those falling further behind:

Value — Is your digital presence organized around the outcomes prospective students are trying to achieve, or around the institutional categories that make sense internally? The shift from the latter to the former is what makes your content useful to an AI model trying to answer a student's question, and useful to the student when they arrive.

Voice — Are your digital decisions grounded in what prospective students actually say, ask, and need — or in what administrators and faculty believe students want? The institutions that are winning the AI-mediated recruitment moment are doing so because they understand the actual language, concerns, and decision criteria of their applicants. That understanding comes from research, not from analytics alone.

Velocity — Does your digital team have the ability to act on what the research reveals? Most university digital programs are structured for governance, not iteration. The shift AI is producing in student search behavior is moving faster than annual website refresh cycles. The institutions that can adapt in weeks rather than years will capture the window that is opening now.

Vision — Do enrollment, digital, communications, and academic leadership share a single, evidence-based picture of the prospective student's journey? In most institutions, these functions are solving different problems with different data. The AI transition requires a unified direction — one map, one strategy, one measurement framework — that aligns every team investing in digital around the same student experience.

The Institutions That Act Now Have a Meaningful Advantage

The window is not permanently open. As AI-mediated search becomes the norm rather than the exception, the institutions that have restructured their digital presence around the student decision journey will build the kind of domain authority — in AI models' characterizations of them, in the quality of student experience when they arrive, in the outcomes they can demonstrate — that is difficult for later movers to displace.

The institutions that wait will not just lose traffic. They will lose the ability to show up at all in the research conversations that shape an applicant's shortlist before any website visit happens. And recovering from that kind of invisible position is a longer, more expensive project than investing in the research now.

The website modernization work and customer journey mapping that sets institutions up to navigate this shift are not multi-year transformation programs. They are a few weeks of rigorous research with real prospective students, producing development-ready direction your team can act on immediately — a current-state picture of how students actually navigate your digital presence, a future-state map of the experience that would serve them well in an AI-first research environment, and a service blueprint that tells you exactly what has to change, in what order, to get there.

That is the work that protects your enrollment strategy in the years when AI-mediated search becomes the default — not a theoretical future, but the present that is already showing up in your traffic data.

That is the work PH1 was built to do. Fourteen years of digital transformation research for organizations investing in the decisions that matter, with the behavioral evidence and senior strategy expertise to move from research to action in weeks, not quarters.

Book a Free Consultation →

Sources

1. Google, AI Overviews and Search — What's New, March 2025.

2. Product Impact Podcast, Episode 51: Agents Will Disrupt Search & Shopping (Devi Parikh, CEO Yutori, ex Meta).

3. Product Impact Podcast, Episode 42: HubSpot's Head of AI on How AI Rewrites Customer Acquisition & Marketing (Nicholas Holland).