ChatGPT
Claude
Perplexity
Gemini
GrokWhat an AI agent for university admissions actually does
An AI agent for university admissions is software that answers applicant questions from your own content and then takes the next step: it qualifies the person, books an open day, records the conversation in your CRM and hands over to an adviser when a human is needed. A chatbot stops at the answer. An agent carries on to the action.
That distinction matters for an admissions office because the costly gaps rarely sit in the answer itself. They sit in the booking nobody made, the enquiry that waited over a weekend, and the applicant who never received a reminder about a missing reference. This article walks through seven concrete use cases, shows where your advisers should take over, and explains what to check in a UK context. For the formal definition and limits, see our piece on the AI admissions agent, its missions and its boundaries.
Use case one: answering course and entry requirement questions
The agent answers questions about courses, entry requirements, accommodation and deadlines using only the content you have approved, and says plainly when it does not know. Applicants ask the same things repeatedly: whether a course accepts a particular combination of A-levels, what the contextual offer policy is, whether a foundation year exists.
The value lies in the grounding. An agent that answers from your course pages and prospectus repeats what you have published, rather than improvising. Where the content is silent or ambiguous, the right behaviour is to say so and offer a conversation with an adviser. Staff then review the questions the agent could not answer, which often reveals gaps in the website itself.
Use case two: qualifying the applicant and routing to the right course
The agent asks a few natural questions, such as level of study, subject interest, home or international status and intended start year, then points the person to the relevant course and the right team. Qualification here means understanding what the applicant needs, not scoring them.
A prospective student asking about business might be interested in an undergraduate degree, a degree apprenticeship or a postgraduate taught course. The agent separates those paths early, so the record that reaches your team already says which one. Home and international applicants follow different routes and deadlines, and routing them correctly saves a first call that would otherwise be spent on discovery.
Use case three: booking open days and campus visits
The agent shows available open day sessions and campus tours, books the applicant in and confirms the details. It can also offer a one-to-one appointment with an adviser or a department contact when the person wants something more specific than a group event.
This is often the most tangible early win, because the booking happens in the conversation instead of after a form and two emails. Applicants who want to know about accommodation or a particular department can be routed to the matching session. Your events team keeps control of capacity and content, and the agent works from the sessions you have opened.
Use case four: chasing missing documents and explaining next steps
After an application is submitted, the agent can explain what happens next and remind applicants about documents still outstanding, such as references, transcripts or evidence of English language ability. It works from the status your team maintains, and it does not guess.
Most UK undergraduates apply through UCAS, so the main application sits outside your own systems. The agent therefore concentrates on what your institution controls: supplementary forms, portfolios, interview scheduling, and the information applicants need when they hold an offer. Clear wording about who holds which part of the process spares applicants from contacting the wrong office.
Use case five: multilingual and out-of-hours answers for international applicants
The agent answers in the applicant's language at any hour, which suits international recruitment across time zones. A prospective student in another continent can ask about English language requirements or the visa support process at midday their time, which may be the middle of the night in the UK.
Test the languages you really recruit in, and check how the agent behaves on sensitive topics. It can point to your published guidance on visas and sponsorship, but it must not interpret immigration rules or promise an outcome. Those questions go to your international admissions team, who know the current position.
Use case six: writing the summary to the CRM and handing over
When a conversation needs a person, the agent passes it on with a summary, so the adviser does not start from zero. It writes the record to the school's CRM: who the applicant is, what they asked, which course they are considering and what they agreed to.
A good handover changes the quality of the first human conversation. The adviser sees the history, the course of interest and the open questions, and can open with something useful instead of repeating the basics. Agree in advance what triggers a handover: a request for a person, a complaint, a disability or welfare disclosure, or a question the agent cannot answer with confidence.
Use case seven: re-engaging a stalled applicant
The agent can follow up with an applicant who asked a question, booked nothing and went quiet, for example by offering an open day or a reminder of a deadline. These are illustrative workflows, and each one depends on the consent the applicant gave and the channels you have permission to use.
Keep the message useful and brief. A reminder that registration for an open day is still available is more welcome than a general prompt. If you add WhatsApp, available as an optional module with Skolbot Chat, check that the applicant has opted in to that channel before any message is sent.
The seven use cases at a glance
The table summarises where the agent acts, where a person steps in and what deserves a check before launch.
| Use case | What the agent does | Where a human takes over | What to check |
|---|---|---|---|
| Course and entry questions | Answers from approved content | Contextual offers, exceptions, unclear cases | Are answers traceable to your pages? |
| Qualifying and routing | Asks a few questions, points to a course and team | Complex academic backgrounds | Are home and international routes separated? |
| Open days and visits | Shows sessions, books, confirms | Accessibility needs, bespoke visits | Does capacity match your events system? |
| Documents and next steps | Explains steps, reminds about missing items | Disputes over what was received | Is the status data up to date? |
| Multilingual, out of hours | Answers in the applicant's language | Visas, sponsorship, welfare matters | Are tested languages the ones you recruit in? |
| CRM summary and handover | Writes the record, passes the conversation | Everything that needs judgement | Which fields are written and with what consent? |
| Re-engaging applicants | Sends a relevant follow-up | Replies that turn into a real conversation | Is consent recorded for each channel? |
What an AI agent must never do
An admissions agent must not decide who gets an offer, and it must not promise anything you have not approved. Decisions stay with your admissions staff, because they involve judgement, fairness and accountability.
Specifically, the agent should never:
- make or hint at an admissions decision, or predict whether an applicant will receive an offer;
- promise a grade requirement, a visa outcome, a scholarship or a bursary;
- invent fees, dates, course details or accommodation availability;
- give immigration, legal or financial advice;
- continue a conversation when an applicant discloses a welfare concern and needs a person.
Build these rules into the configuration and test them with awkward questions before launch. A useful exercise is to have experienced advisers try to push the agent into promising a place.
UK context: UCAS, Clearing and regulation
In the UK, the agent has to fit the admissions calendar. Enquiry patterns cluster around open days, the main application deadline and, later, Clearing. During Clearing, course availability changes quickly, so the agent should say what is confirmed and send anything uncertain to the team. Apply the same care to Adjustment and to UCAS Extra, where applicable.
On regulation, three bodies are worth knowing. The Office for Students regulates registered providers in England and expects accurate information for prospective students. QAA is the reference for quality standards across UK higher education. The ICO oversees UK GDPR and the Data Protection Act 2018, so you need a lawful basis for processing chat data, a privacy notice reachable from the conversation and a processing agreement with your provider.
Tell applicants they are talking to an automated assistant. It is good practice, it builds trust and it prepares you for transparency expectations that are growing across Europe. Russell Group universities, post-92 institutions and independent colleges all face the same basic duty: what the agent says about a course must match what you can stand behind.
How to start with a thirty-day pilot
Start with one course area and two actions, then judge the result on your own terms. A narrow pilot tells you more than a broad launch, because advisers can read every conversation.
- Week one: choose a department, load approved content and agree the handover triggers.
- Week two: advisers test the agent with common and difficult questions, and you fix gaps in the content.
- Weeks three and four: go live on selected pages, review conversations weekly and note unanswered questions, handovers and bookings.
Before you begin, write down what success looks like, for example fewer unanswered out-of-hours enquiries or cleaner CRM records. For a wider view of other scenarios, see our guide to conversational AI use cases beyond admissions and our article on chatbot scenarios that increase enrolment. The full picture is in the guide to AI chatbots for student recruitment.
Skolbot Chat is a website agent that answers from your institution's own content, qualifies applicants, books open days and appointments and pushes the record to your CRM. WhatsApp is an optional module.
