Chatbots for schools, course creators and online education

Late July is the calm before the wave. In a few weeks, everyone who has been "thinking about" your September course will decide at once. They'll ask whether there are still places, whether they can pay in instalments, whether the evening group clashes with work, and why the login link in their welcome email doesn't work. Most of these questions will arrive on a Sunday night, and many of them will be asked three times on three channels.

Schools, language academies, training providers and online course creators all share this pattern. Support isn't steady; it comes in waves around terms, cohort launches and deadlines, and small teams get buried exactly when enrolments are on the line. This guide shows how to use chat and AI to answer the predictable questions instantly and keep your people for the conversations that need them.

Why education support breaks at peak time

• Everything happens at once. Enrolment, payment, timetables and access questions all peak in the same two or three weeks.
• Prospects compare options. Someone choosing between three courses goes with the provider that answers clearly and quickly.
• Students study at night. Access problems at 11pm before a deadline feel urgent, and email replies the next afternoon feel like no reply.
• The same answers, all day. Start dates, prices, entry requirements and refund rules, typed again and again.
• Sensitive topics mixed in. A question about a missed deadline may be about illness or money, and needs a person with care.

A good setup separates the predictable from the personal, and handles each properly.

What students and prospects ask

Question	Best channel	Automate?
Course dates, times, format, prices	Website chat	Yes, from course pages
Entry requirements, levels, placement	Website chat, email	Partly: answer, then a placement form or call
Payment plans, invoices, discounts	Website chat, email	Partly: answer the policy; staff handle individual cases
"I can't log in" / access problems	Chat, email	Yes for common fixes; hand over if it persists
Deadlines and extensions	Email	No: a person decides
Certificates and transcripts	Email	Partly: explain the process; staff issue documents
Refunds and cancellations	Email	Partly: explain the policy; staff decide
Complaints, wellbeing, accessibility needs	Email, phone	No: a person, quickly

Step 1: Fix the content before the wave

An AI agent repeats what your pages say. Before enrolment opens, make sure each course page states dates, schedule, format, price, payment options, entry requirements and what is included. Add a short FAQ for access, refunds and certificates. If last year's dates are still on a page somewhere, remove them.

Step 2: Answer the predictable questions with AI

Train an AI agent on your course pages, FAQ and policies, and put it on your website chat and messaging channels. Introduce it honestly ("I'm the AI assistant; I can answer most questions about our courses, or get you to the team"), and test it with 30 real questions from last year's inbox before the peak.

For an introduction to the difference between rule-based bots and AI agents, read how to choose an AI agent for customer service.

Step 3: Collect enrolment details with a form

Chat is great for questions and poor for collecting ten fields of personal details. When someone is ready to apply, send them to a proper enrolment or placement form, with logic for different courses and levels and payment where possible. Keep the chat for the questions that come before and after.

If a prospect asks three questions, they're interested. Send the enrolment link with the third answer.

Step 4: Route the human conversations

Set clear handoff rules so the right person picks up:

• Admissions: placement, special entry cases, group bookings.
• Finance: individual payment plans, invoices, refunds.
• Academic staff: deadlines, extensions, course content questions the AI can't answer.
• Student support: wellbeing, accessibility and complaints, always to a person and quickly.

Make sure the person sees the whole conversation so the student never repeats a difficult story.

Step 5: Plan for the peaks

• Extend hours for two weeks around enrolment, not all year.
• Send proactive messages: welcome emails with access instructions, reminders before start dates and deadlines.
• Post known issues (a login outage, a delayed timetable) in the chat greeting, so the AI and staff answer consistently.

Student data and safety

• Keep records out of chat. Don't let a chat tool see or share grades, transcripts or personal circumstances. Handle those in your student system.
• Be clear about AI. Say when someone is talking to an AI assistant and how to reach a person.
• Take extra care with children. If your learners are under 18, data protection rules are stricter. In the US, student education records are protected by FERPA; in the EU, GDPR applies. This isn't legal advice, so check the rules for your learners and country.

Common mistakes

• Letting the bot give academic advice. It should explain options, not recommend a path for someone's career.
• Enrolment by chat interview. Twenty questions in a chat window lose applicants. Use a form.
• Old dates on old pages. The AI will find them and quote them.
• No plan for the wave. The week a cohort starts is not the week to set up support.

A setup before the September wave

1. This week: export last year's enrolment-season questions and list the top 20.
2. Next week: update course pages and the FAQ so each question has a clear answer.
3. Early August: add website chat with an AI agent, connect WhatsApp and email, and test with real questions.
4. Mid-August: set up the enrolment form, handoff rules and proactive welcome messages.
5. From enrolment day: read AI transcripts daily and fix content gaps the same day.

For most small schools, academies and course creators, Tidio covers chat, AI answers and the shared inbox in one tool, alongside a form tool for applications. To compare more options, browse the live chat category, or read the difference between an AI agent, a chatbot and a copilot.