This essay is excerpted from a new Chronicle special report, “The Home Stretch of Student Recruitment,” available in the Chronicle Store.
You’ve just wrapped up an event for admitted students. The list of attendees is in front of you and now comes the part that matters most: making sure they show up on your campus in the fall. But with so many students to reach, and limited time, it’s easy to feel stuck and overwhelmed. That’s where generative AI can help: Not by replacing your expertise, but by acting as a time-saving collaborator on routine enrollment tasks so you can focus on making personal and authentic connections with students and parents.
In my work as a higher-ed-enrollment marketer, I’ve spent years studying yield — the share of accepted applicants who commit to a college. In a recent series of my podcast, The Application, I explored how AI can enhance “yield season,” that anxious period between when admissions decisions go out and when students commit to a campus. I’ve thought a lot about what drives student decision-making and even wrote my dissertation on audience-centered marketing in higher education. So when I say AI can help with yield, I mean I’ve tried it myself.
What I’ve learned is that AI doesn’t have to make our work feel less personal. Used thoughtfully, it can actually do the opposite.
We’re in a moment when every new tool claims to be “powered by AI.” I get at least one email from a vendor every day with that pitch. At the same time, many enrollment professionals are skeptical about using generative AI in their work. They’re worried about ethics, privacy, even the environmental impact. Others simply haven’t had the time to explore AI and don’t know where to start.
Those reservations are valid and worthy of real conversation. We should be asking hard questions and making deliberate choices. My aim here is to make a different case: That AI can actually support the human side of our work. That these tools, used with intention, can free up our time to do the parts of our work that matter most — like connecting with students, providing them with clear and relevant information, and showing we understand them.
Think of AI as a tech-driven intern, not a substitute for your knowledge.
Skepticism about AI’s ability to do any of that is healthy and forces us to ask: What does this tech actually solve in the admissions context? And what should it never touch?
In yield season, one of our biggest challenges is scale. Students and parents expect customized and personalized information to guide them in making this choice. That is simply how we all experience life as consumers now. AI can help us meet these expectations more efficiently and effectively.
Think of AI as a tech-driven intern, not a substitute for your knowledge. It can help you move faster, but you still need to steer. For example:
- A student submits an inquiry form on your campus website asking about their intended field of study. Halda, an AI-driven tool, instantly serves up content tailored to their interests without endless clicking or digging through your site. The student gets what they’re looking for immediately, and you get an insight into that student. You now know not just their prospective major but what they most want to know about it. When you reach out to the student, you’re not starting from zero, and you can provide a relevant, truly personal outreach.
- Or take personalized videos. Allied Pixel creates dynamic, data-driven videos that adapt to each student’s name, academic interests, location, and even financial-aid status, tailoring the narrative to what matters most to each viewer. For a prospective nursing major from out of state, the video might showcase the university’s lab facilities and dormitories. A local business major could see highlights of commuter support and nearby internship opportunities.
- Element451 offers AI-powered agents that help students navigate financial aid, stay on top of deadlines, and connect with campus resources, supporting both yield and retention. These AI agents can also help students stay engaged after committing to college and making their deposit, which is an increasingly critical step for reducing melt.
- On my podcast, Kelsey Smith and James Cox, both of Purdue University’s business school, explained how they use progressive profiling to collect information from prospective students gradually, and from multiple sources, instead of overwhelming them with long forms to fill out up front. As students interact with the admissions office, their profiles grow richer. The role of AI here is to help the admissions team tailor its responses: Someone focused on affordability might see messaging about the value and return on investment of a bachelor’s degree, while a student asking about campus life might receive content highlighting student experiences. As Kelsey put it, “They’re getting similar information — it’s just presented in different ways to get them to take the action that’s best for them.”
- Another compelling example comes from the University of California at Davis. On the AI for U podcast, Lisa Nguyen, a digital marketer supporting the university’s part-time M.B.A. program, shared how her team uses ChatGPT to analyze their own application data. With enrollments declining, they needed to pinpoint the most effective windows for marketing their programs. Historically, they relied on anecdotal input from program directors to determine timing. Now, using AI, they’ve been able to identify trends — such as when applications are most likely to be submitted — and adjust their marketing accordingly. For prospective students that means they’re more likely to hear about the program at just the right time — when they’re thinking about applying but haven’t yet made a decision. That kind of well-timed, relevant communication can improve engagement and, ultimately, increase yield.
Those are creative, ambitious ideas for how AI tools can help us bring more of the students we admit to the campus, come fall. But what would everyday use of AI look like in our roles? Below are a few ways that enrollment and marketing teams can put generative AI to work — starting with simple, low-risk tasks that support personalization without sacrificing authenticity.
Use it to prepare for one-on-one conversations with prospects. Imagine using ChatGPT, Bard, or another generative-AI platform to summarize what you know about a student before a call: the major they plan to pursue, the prospective-student events they’ve attended, the questions they’ve asked. The AI gives you talking points. You spend less time digging through notes and more time building personal connections with the student.
Ask AI to help you tailor a message to different groups of recipients. Take a routine task like sending out deposit reminders to admitted students. AI can help you write multiple versions of the reminder email — one for STEM students, one for first-generation college students, and one for students who live in counties near the campus. Small shifts in tone and relevance can mean big changes in response rates. When students receive content that reflects their interests and background, they “feel seen” by the college.
Still skeptical? Start small. Use an AI tool to help brainstorm the subject line of a marketing email. Ask it to help you update an old blog post. Or, if your college is already using a chatbot to capture the concerns of prospective students, ask AI to analyze the transcripts for trends. You might discover housing is the top concern this month, or that admitted students are asking more questions about career outcomes than in previous years. With those insights, you can update your messaging, adjust your outreach strategy, or create content that directly responds to those concerns.
AI can also help your marketing staff with A/B testing (comparing two variations of content to see which is more effective). Create two versions of an email: one written entirely by you and one written with the help of AI. Compare click-through rates, or see which version earns more clicks or replies.
Here’s the line I wouldn’t cross. I would never use AI to generate entirely original content meant to sound like a human wrote it, especially if you’re not reviewing or editing it carefully. Generic messages can’t replace genuine voice or real insight, and they do not help us to differentiate our institutions. A post written entirely by ChatGPT for a new-students blog might look clean, but students are smart. They can tell when something doesn’t sound like it came from a real person or reflect the actual tone of your institution.
Again, the goal is to use AI to assist your work, not to replace your work.
This very essay is an illustration of what I mean: I wrote it, but I also used generative AI in the outlining and editing stages of the process. An AI tool helped me structure an outline for my ideas, and once I’d written a draft, it pointed to unclear sentences and made proofreading suggestions. Every idea, decision, and example I’ve shared here came from real conversations, real experiences, and a human who still believes that authentically connecting with students and families is at the heart of admissions work.
We shouldn’t use AI to replace creative thinking or strategic storytelling. AI doesn’t know your campus culture. It can’t feel what a student is feeling the week before their deposit is due. But it can make the work behind personalization — like segmenting, repurposing, and analyzing — much faster and more manageable. In fact, the more time AI saves you, the more time you can spend on what matters most: building relationships with students and families.
