AI Will Destroy University Website Traffic. There's Only One Path Forward.

2026 was already a difficult year for universities. Financial pressure. Enrollment volatility. Intensifying competition. Now add this: the traffic you depend on is being restructured by AI — faster than most institutions are prepared to respond.

By the second half of 2026, Google is replacing the majority of its traditional search results with an AI Mode that synthesizes answers and delivers them directly on the results page. The prospective student gets their answer. They never visit your website.

The most challenging part: there is no guarantee your institution is the source of that answer — or that the answer is accurate.

The Risk to Universities and Colleges

Google is systematically cannibalizing your website's content and audience — replacing your pages as the destination with AI summaries that answer prospective student questions without a single visit.

• 93% of Google AI Mode sessions end without a click to any website (Semrush, 2025)

• 58% fewer clicks on top-ranked pages when an AI Overview is present (Ahrefs, February 2026)

• 99.9% of informational search queries now trigger AI-generated answers — the category covering virtually everything a prospective student asks

• Gartner forecast a 25% decline in traditional search traffic by 2026 — and called it conservative

• Science and education queries see AI Overview triggers on up to 43% of searches — among the highest exposure of any industry, compared to just 3% in e-commerce (Ahrefs data via Pasquale Pillitteri, 2026)

The applicant journey built on organic search is contracting, regardless of how well your team executes on content or SEO.

University enrollment marketing has not faced a structural disruption of this scale before. Most institutions have not started responding.

What Students Actually Search For

• Program requirements and prerequisites — "What do I need to get into [program]?"

• Tuition and financial aid options — "How much does [university] cost per year?"

• Career outcomes and graduate employment — "What jobs do [university] graduates get?"

• Application deadlines and process — "When is the application deadline for [university]?"

• Campus life and student experience — "What is it actually like to study at [university]?"

That's what students had been traditionally searching for. The issue is that in reality LLMs and heavy use of Reddit is shifting their behaviour towards deeper, more personalized questions.

Applicants are asking more complex questions because university websites so poorly enable them to answer the questions about their $100,000 decision of where to study. On most university websites even answering simple questions like "what will my degree cost me" is difficult and same goes for "which classes would help me improve my chances of getting accepted into the business school."

LLMs and Reddit are cannibalizing these high-intent search queries because of inaction to reorient university websites towards how applicants actually seek information.

Institutional Silos Are Destroying Website Performance

Most university domains are case studies in how to rank poorly: fragmented websites, duplicate and contradictory information, pages that haven't been updated in years, and no clear next steps for users.

The issue is not Google or AI. It is the institutional silos that undermine website performance. Google and AI systems optimize for the most authoritative and efficient path to answer a user's question. Universities should be among the most authoritative sources online, yet they dilute their own authority by fragmenting information across central, faculty, departmental, and program-level websites.

Here's an example of how the typical university website architecture is fragmented:

• Registrar teams control admissions requirements, but often cannot answer program-specific questions.

• Faculties control degree information, but often cannot answer questions about career outcomes.

• Departments control course and curriculum information, but often cannot answer admissions-related questions.

The solution is not more content — it is better information architecture. Institutions that consolidate fragmented web presences into a coherent, structured information system can resolve both the AI citation problem and the broader digital transformation debt these silos create. Accelerating that consolidation may be the highest-leverage investment a university can make to protect the integrity of the applicant journey.

Understanding the Applicant Journey

University websites need to be viewed as products that solve the problems of applicants, not as a digital brochure. This comes down to structuring websites around the unique needs at each stage of the applicant experience.

The applicant experience is a research process that spans months, crosses multiple channels, and involves a prospective student — often alongside their family — evaluating whether your institution is worth a commitment of four years and a massive investment. At every step in that journey an applicant is trying to complete a task and will increasingly decide whether to use AI or your website.

Unfortunately our research has found that poor university website experiences and vague, inconsistent answers have pushed more to rely on LLMs.

A prospective student in 2026 typically:

1. Asks ChatGPT or Claude a broad question about programs or careers

2. Uses Reddit or TikTok for unfiltered student perspectives

3. Runs a Google search — and receives an AI-generated summary

4. Visits your website only if that summary raised a question the AI couldn't answer

Most enrollment marketing teams are optimizing for a journey that no longer reflects how prospective students actually move from curiosity to application.

Mapping the AI-era applicant journey — understanding where AI intercepts it, where institutional touchpoints still hold weight, and where new touchpoints need to be created — is the foundational diagnostic your strategy depends on. You cannot redesign a funnel you have not accurately mapped.

Nurturing Relationships Is the Product

Google will continue to cannibalize your content pages. That process is not reversible. But it cannot cannibalize the relationships your institution has with prospective students who have already chosen to engage directly.

• A contact is a business outcome. Every email, campus visit request, open house registration, and information inquiry is a prospective student who bypassed AI and chose your institution directly.

• AI makes contact management more valuable, not less. Institutions that deploy AI for faster inquiry response, personalized follow-up, and CRM-driven nurturing are deepening relationships that SEO alone never could.

• Technology enables the shift from broadcast to dialogue. Automated but personalized communication sequences are now accessible to institutions of any size — the operational advantage is available to those who build for it.

• Every contact is a direct channel immune to algorithm changes. A prospective student who has opted into your communications is no longer subject to what Google decides to show them next.

The product is not the website page. It is the tools that develop the relationships. 2U research with university partners found that when program content was structured for AI discoverability, ChatGPT-driven traffic converted at twice the rate of other organic sources — prospective students arrived further along in their decision-making process. Build the infrastructure to capture that moment. See AI Strategy for Universities and Colleges.

Focus on Information Architecture and Navigation

The question is not whether your website has enough content. Most universities have too much — spread across too many owners, in too many formats, organized around departmental logic rather than applicant logic.

What poor information architecture costs you:

• Contradictory facts across departments — when registrar, marketing, and faculty pages say different things, AI surfaces the contradiction or defaults to an aggregator it trusts more

• Navigation built for staff, not students — menus organized by internal structure rather than by the questions applicants actually ask

• PDF-heavy content — AI cannot reliably extract or cite content locked in PDFs; neither can prospective students on mobile

• No content hierarchy — AI cannot determine which institutional page is authoritative versus supplementary when dozens address the same topic

What strong information architecture delivers:

• Question-first page design — every program page structured around what applicants actually ask, not what departments want to publish

• A single source of truth for every critical fact, with a named owner accountable for accuracy and recency

• Schema markup built into the structure from the start, not retrofitted — deadlines, tuition, requirements tagged with structured data so AI cites you, not an aggregator

• Clear paths from any entry point to application — navigation that treats every page as a potential first touchpoint in the enrollment funnel

Seer Interactive research found that brands cited in AI Overviews see a 35% higher clickthrough rate than those not cited. That citation advantage starts with architecture, not content volume.

Build Unique Content and Resources

The applicant making a $100,000 decision needs more than a brochure. They need a website that actually helps them complete the tasks they came to do.

AI cannot replicate what only your institution holds — but the goal is not to produce more content. It is to build the tools and resources that answer the questions applicants cannot get answered anywhere else:

• Program-specific outcome data — employment rates, graduate salary ranges, and time-to-placement by cohort that no aggregator can fabricate

• Total cost calculators — not tuition tables, but tools that let an applicant model their actual cost of attendance including financial aid eligibility

• Student and alumni accounts — first-person video and written stories that reflect the specific experience at your institution, not a synthesized description

• Admission pathway tools — resources that help applicants understand exactly what their application path looks like given their background

A prospective student who can complete their research tasks on your website has no reason to take their questions — and their attention — to an LLM.

Become the Best University Answer Engine on the Internet

Ask your website what a degree will cost over four years, which programs have the best employment outcomes, or what GPA a student actually needs to get in. If the answer is a departmental menu, a PDF, or a page last updated two years ago, the website is functioning as a digital brochure — not as the answer engine your prospective students need it to be.

The shift requires information architecture, not more content:

• Every program page structured around applicant questions, not departmental descriptions

• FAQ schema markup on every question-answer pair so AI cites you as the source of record

• Decision-support tools that require direct engagement — eligibility checkers, cost models, application planners

• Structured data that makes institutional facts machine-readable, not just human-readable

Google's AI Mode runs 16 parallel sub-searches per query. Structured institutions get cited. Unstructured ones get paraphrased without attribution.

The goal is not to publish more content. It is to make your institution the definitive answer source for every question a prospective student brings to the decision. See Website Architecture Modernization.

Make Your Website the Source AI Points To

The institutions that neutralize this risk share a common approach: they stop competing for clicks and start competing for citation authority.

When your website becomes the most structured, most accurate, most frequently updated source of information about your programs, outcomes, and institution, AI systems cite you by default. Prospective students who encounter your institution in an AI response arrive pre-qualified and further along in their decision. 2U research with university partners found that ChatGPT-driven traffic converted at twice the rate of other organic sources.

What winning institutions build:

• Tools that keep visitors onsite — cost calculators, program finders, and admission pathway tools that answer questions directly rather than redirecting to another department

• Internal search that actually works — the fastest path to losing a prospective student is a site search result that returns irrelevant pages or surfaces outdated content

• Proprietary data no aggregator holds — graduate outcomes, salary by cohort, time-to-completion, structured so AI has no better source to cite

• Direct relationships before Google Zero arrives — email programs, campus visits, open house conversion — contacts that remain intact regardless of what Google changes next

The Chegg case is the cautionary version of this story. Chegg sued Google in February 2025 when AI Overviews absorbed its content, reporting a 24% year-over-year revenue decline before the lawsuit was filed. Chegg made the same mistake most universities are making: treating a website as a passive content source rather than an authoritative answer engine. Publishers call it Google Zero. Your institution does not have to follow the same path.

5 Steps to Do Immediately

1. Map your current applicant journey end-to-end. Chart every touchpoint from first AI query to application submission. Identify where the enrollment funnel is being intercepted and where direct contact-building needs to start. See Applicant and Student Journey Mapping.

2. Audit your website information architecture — not content volume, but structure. Identify every page that duplicates, contradicts, or fragments a critical institutional fact. Map ownership across departments. The audit surfaces the consolidation work required before any AI strategy can function. See Website Architecture Modernization.

3. Test each major page against the task an applicant would bring to it. Not "does this page exist" — but "can a prospective student answer their question here, without clicking elsewhere or calling admissions?" Pages that fail this test are the ones AI has already replaced.

4. Implement schema markup across all program and admission pages. Once architecture is consolidated, structure key facts as machine-readable data so AI cites your institution directly rather than a third-party aggregator. See Benchmark Institutional Performance.

5. Have an honest conversation about your internal search. Most university site search tools return irrelevant results, surface outdated pages, and fail to answer the questions applicants actually bring. Every poor result sends a visitor to Google or an LLM instead. Internal search performance is a direct diagnostic of your content strategy and information architecture — the same fixes that improve it improve your AI citation authority at the same time.

The PH1 Advantage for Universities

PH1 has led digital transformation and product research for over a dozen universities, and brought the same discipline to Spotify, Microsoft, Mozilla, the National Football League, La Liga, and Hims & Hers. The work is the same: map how people actually move through a decision, find where the experience breaks down, and fix the architecture — not just the content.

For universities navigating this transition, PH1 typically starts with three engagements:

• Applicant Journey Mapping — charting the AI-era journey from first query to application, identifying every intercept point and gap

• Website Architecture Modernization — auditing and restructuring your information architecture for citation authority and direct engagement

• AI Strategy for Universities and Colleges — building the strategic plan for how your institution competes for prospective students in an AI-first discovery environment

The institutions that act in 2026 will own the prospective student relationship.

Book a Free Consultation →