How to Modernize the University Applicant Experience for the AI Era

The Applicant Research Process Has Already Changed — Here Is How to Respond

If you are responsible for enrollment, digital strategy, or communications at a university or college, you have likely noticed something shifting in your traffic data over the last year. Organic visits to program pages are softening. The search terms that once reliably drove prospective students to your website are producing fewer clicks. And prospective students, when they do arrive, seem to already have opinions about your institution that they did not form on your site.

You are not imagining this. The research process that prospective students follow — from initial interest through shortlisting through application — has structurally changed. And understanding how it changed is the first step toward positioning your institution to thrive in the environment that is replacing it.

An EAB national survey conducted in late 2025 found that 46% of high school students now use AI tools — ChatGPT, Gemini, Perplexity — during their college search, nearly double the 26% who reported doing so just six months earlier. Everspring's 2025 AI Search Trends Report, drawing on 450,000+ real student interactions, found that two-thirds of prospective students now use AI tools before turning to Google. Among 18-to-24-year-olds, nearly half use ChatGPT as their first stop — ahead of any traditional search engine.

This is not a trend that will reverse. It is the new baseline. And it is reshaping what prospective students expect from the institutions they are evaluating.

How University Websites Got Here — and Why It Matters Now

The digital model most universities still operate on is a faithful translation of the admissions guide into website form. A prospective student arrives, navigates through institutional pages organized by faculty and program, finds information relevant to their situation, and reaches out to an advisor when they need help. The website is a self-serve information architecture — comprehensive, authoritative, and built around the institution's internal structure.

This model made sense for a long time. Prospective students arrived with intent — they came for the institution and often with a particular program in mind. The website's job was to confirm a decision they were already leaning toward and move them toward an application.

When search became the dominant way people found answers, universities adapted. Faculties and programs added pages that answered institutional questions from their own context — financial aid pages under each faculty, student life content under each program, career outcome pages scoped to individual departments. The intent was to capture search traffic from students researching specific questions.

The result was predictable. Content proliferated. Pages duplicated across faculties with slight variations and different update cycles. Outdated pages lingered deep in the site structure, inaccessible through navigation but still indexed by search engines. The institutional website became a sprawling, layered artifact that reflected decades of accumulated organizational decisions rather than any coherent student experience.

It was convoluted — but functional. Advisors and recruiters knew where to find the answers and where to direct applicants. The system worked as long as a human was the bridge between the student and the information.

That bridge is no longer the default path. And the model it supported needs to evolve.

How AI Is Changing What Prospective Students Expect

The shift is not just about where students start their research. It is about what they expect the experience to feel like.

A prospective student who has spent the last year interacting with ChatGPT, Claude, and Perplexity has internalized a new standard for how information should work. They expect answers tailored to their specific situation — not generic program pages they have to interpret themselves. They expect synthesis — not a maze of links across faculties and departments. They expect to be supported through a decision, not handed a filing cabinet and told to browse.

As we explored in AI Will Destroy Your University Website Traffic, Gartner predicted that traditional search engine volume will drop 25% by 2026 due to AI chatbots and virtual agents. Seer Interactive found that organic click-through rates dropped 61% for queries where Google AI Overviews appear. For universities that have built their entire digital acquisition model around organic search, these are not marginal adjustments — they are structural reductions in the primary channel.

But the deeper challenge is what happens when students take their most important questions to AI — and AI answers them from whatever sources it can find.

Students are asking questions institutions cannot answer with certainty. What kind of job should I expect with this degree? What are my chances of getting admitted with my transcript? Will this program prepare me for a career in a field that may not exist yet? These are not questions an admissions team can answer with precision. They depend on labor markets, individual circumstances, and outcomes no institution can guarantee. But AI is answering them — drawing from program review sites, Reddit threads, outdated rankings, and commentary from people who may never have set foot on your campus.

The EAB survey found that 18% of students removed a college from consideration based on information surfaced through AI-generated search results. Students are making enrollment decisions based on AI-synthesized information — and institutions have limited visibility into what those answers are saying about them.

Sarah Gold, founder of Projects by IF and one of the most respected voices on trust in digital services, argued on the Product Impact Podcast that trust in AI-mediated services is fragile and asymmetric — easy to lose, almost impossible to recover once broken. A prospective student who receives a confidently wrong answer about your institution from an AI assistant may never visit your website to verify it. The misinformation becomes their reality.

This is not a reason to panic. It is a reason to act — because the institutions that solve this problem will earn a kind of trust and visibility that was never available through traditional search alone.

Universities Rising to the Challenge

The encouraging reality is that institutions across the world are already demonstrating what is possible when they invest in meeting students where they are.

Georgia State University deployed an AI-powered advising chatbot and reduced summer melt — the gap between acceptance and enrollment — by more than 50%, while achieving a 5% increase in retention and a 3% increase in re-enrollment. The investment worked because it met students in the moments that mattered: the anxious weeks between acceptance and arrival, when questions go unanswered and commitments quietly unravel.

Arizona State University became the first US university to formalize a partnership with OpenAI, giving its entire campus access to ChatGPT Enterprise. Within weeks, the institution received hundreds of proposals from faculty and staff, ultimately launching 200+ active AI initiatives across teaching, research, and administration. The scale of adoption demonstrated something important: when the institutional framework supports experimentation, the demand for AI-enhanced experiences is already there.

Bolton College in the UK built a platform empowering teachers, students, and administrators to create and share AI chatbots — handling over 1,000 timetable enquiries per day at the start of the academic year, freeing staff to focus on the higher-value interactions that require human judgment.

These are not isolated experiments. EDUCAUSE's 2025 AI Landscape Study found that 57% of institutions now consider AI a strategic priority, up from 49% the year before. The direction is clear. The question for every institution is whether to lead it or follow it.

And BCG's recent research, How AI Can Help Higher Education Capture a Once-in-a-Generation Opportunity, frames the stakes directly: with enrollment projected to drop as much as 15% between 2026 and 2029, AI is not an enhancement — it is the critical lever for institutions navigating the enrollment cliff.

Three Capabilities That Position Your Institution for Success

The institutions that will navigate this transition well share a common approach: they invest in understanding the student experience before they invest in technology. The technology serves the strategy — not the other way around.

1. Reclaim Control of Your Institutional Narrative Through a Knowledge System

The misinformation problem is a content problem at its root. AI models synthesize answers from the information available about your institution across the web. If your authoritative, current, well-structured content is the strongest signal, the AI's characterization of your institution will be accurate. If it is not — if program pages are thin, if outcomes data lives in PDF reports nobody can find, if faculty research strengths are buried in departmental sub-sites — the AI will work with what it has.

The first capability is a unified institutional knowledge system that consolidates the information students actually need — across programs, across faculties, across the full decision journey — into content structured and rich enough to become the dominant source AI models draw from when characterizing your institution. This is answer engine optimization: ensuring that when an AI synthesizes information about you, the answer is accurate, comprehensive, and reflects the experience you actually deliver.

2. Reorganize the Digital Experience Around the Student Decision Journey

As we argued in Customer Journey Mapping Is the Secret to Unlocking AI Value, the starting point for any digital investment is a rigorous understanding of the actual customer journey — what people are trying to accomplish, in what sequence, with what information, at what emotional stakes.

For universities, that means mapping the prospective student's decision journey across every stage — awareness, research, shortlisting, application, acceptance, enrollment — and understanding what information, in what form, serves the student at each moment. The output is not a new website structure guessed from best practices. It is a student-centered information architecture built on behavioral evidence from real prospective students — one that consolidates duplicated content, eliminates the outdated pages lingering in search indexes, and organizes the experience around decisions students are actually making.

3. Design AI-Powered Experiences That Support Students Through Key Decisions

This is where the opportunity becomes genuinely differentiated. Most institutions will spend the next two years debating whether to add a chatbot. The institutions that pull ahead will build something fundamentally different: AI-powered experiences that help prospective students navigate the decisions that matter most — program selection, financial planning, career alignment, fit evaluation — using the institution's own authoritative data, delivered in the personalized format students now expect.

This is not about replacing advisors. It is about extending the reach of advising into the moments when students are researching at midnight, comparing programs on their phone, or trying to understand whether a specific degree leads to the career they are imagining. Georgia State proved that meeting students in those moments changes outcomes. The institutions that build this capability — accurately, helpfully, at scale — will earn the kind of trust that converts research into applications.

The concept requires validation before development. As we explored in why digital transformation projects fail, the most common failure pattern is investing in technology before understanding what users actually need. An AI concept validation process that puts prototype experiences in front of real prospective students is what separates the investment that lands from the pilot that stalls.

How to Build the Shared Vision Your Institution Needs to Move

Anyone who has worked inside a university knows the reality: websites are governed by committee, negotiated across faculties, and constrained by the autonomy that academic departments rightly protect. Agreeing on wording changes to the homepage can take months. The governance model is designed for stability, not speed.

This is often framed as the reason universities cannot move fast on digital. But it is also the reason the opportunity is so large for institutions that find a way forward.

If your peers are also governed by committees that take months to agree on a homepage, the institution that builds a shared vision across faculties — grounded in evidence everyone can see themselves in — creates an advantage that is structurally difficult to replicate. The barrier to entry is not technology. It is organizational alignment.

Peter Merholz, author of Org Design for Design Orgs, argued on the Product Impact Podcast that organizations need to fundamentally rethink how they are structured to succeed with AI — not just retrain existing teams on new tools. For universities, the digital transformation challenge is not a technology decision. It is an organizational design decision: how do you create a shared picture of the student experience that enrollment, communications, academic leadership, and IT can all orient around?

The answer is research. Not the kind that takes a year. The kind that takes weeks: structured behavioral evidence from real prospective students, across segments, across program types, across stages of the decision journey. A current-state map of how students actually experience your institution digitally. A future-state vision of the experience that would serve them in an AI-first research environment. And a service blueprint that shows every faculty and department what needs to change, in what order, with development-ready specificity.

That shared picture — grounded in student evidence, not internal assumptions — is what unsticks committee-based governance. It is hard to argue with what the students themselves told you.

When PH1 conducted research and service design for Simon Fraser University, the process involved 45+ interviews across researchers, faculty, and students — producing a behavioral picture of the jobs people were actually trying to accomplish, and a prioritized sitemap grounded in customer evidence rather than institutional assumption. That is the kind of work that gives an institution the confidence to move — and the direction to move toward.

The Digital Acceleration Pillars — A Framework for University Digital Leaders

PH1 structures every university engagement around four shifts — the Digital Acceleration Pillars — that separate institutions compounding advantage from those losing ground:

Value — Reorganize the digital experience around the outcomes students are trying to achieve. Every page earns its place by serving a specific student decision. Everything else gets consolidated or retired.

Voice — Ground every decision in what prospective students actually say, search for, and struggle with. The evidence from behavioral research gives institutional leadership the confidence to move — because the direction is student-driven, not opinion-driven.

Velocity — Deliver development-ready direction in weeks, not the year-long discovery process institutional governance cycles expect. The website modernization work that positions institutions for the AI era produces results your team can act on immediately.

Vision — Create the single, evidence-based picture of the prospective student journey that enrollment, communications, faculties, and IT all share. One map. One strategy. One set of priorities. This is what turns committee-based governance from a blocker into an asset — because everyone is working from the same student evidence toward the same north star.

The Opportunity Is Now — and It Rewards the Institutions That Move First

AI has changed how students discover, research, and evaluate institutions. That change is not coming — it is here, measurable in traffic data and enrollment behavior. The institutions that invest now in understanding the student decision journey, building content systems that AI models can draw from accurately, and designing AI-powered experiences that support students through key decisions will build a structural advantage that later movers will struggle to close.

Most institutions will spend this window debating. Some will spend it building chatbots without first understanding what students need. A few will do the research that makes every subsequent investment — in content, in AI experiences, in website architecture — defensible, evidence-based, and pointed at the moments that actually move enrollment outcomes.

That is the work PH1 was built to do. We work across all faculties to build a shared vision and plan. We collect evidence from across your core audience groups and student stages to define the north star. We determine how to best leverage AI in the applicant experience and how to improve organizational use of AI to support it. Senior researchers and strategists. Development-ready results in weeks.

Book a Free Consultation →

Sources

1. EAB, AI in College Search Survey, 2025.

2. Inside Higher Ed, Survey: How High Schoolers Are Using AI in College Search, February 2026.

3. Everspring, 2025 AI Higher Ed Search Trends Report, 2025.

4. Gartner, Gartner Predicts Search Engine Volume Will Drop 25% by 2026, February 2024.

5. Seer Interactive, AI Overview Impact on Google CTR, September 2025.

6. Mainstay, Georgia State University: Behavioral Intelligence for Student Retention.

7. BCG, How AI Can Help Higher Education Capture a Once-in-a-Generation Opportunity, 2026.

8. EDUCAUSE, 2025 AI Landscape Study, 2025.

9. Product Impact Podcast, Episode 29: Trust is a Double-edged Sword — AI will Transform Services (Sarah Gold).

10. Product Impact Podcast, Episode 21: Redefining Design — Peter Merholz on AI and Organizational Change for Product Teams.