Skip to content

Conversational AI for education: student enrollment on WhatsApp

How schools and education networks use conversational AI on WhatsApp to answer families right away, log candidates in the CRM and track the enrollment funnel.

Marlos Carmo

Founder of Tolky

Updated on November 16, 2026

9 min read
Conversational AI for education: student enrollment on WhatsApp

Share

During enrollment season, a school's front office receives more messages than its team can handle. The questions are almost always the same, and families who do not get an answer in time choose another institution.

The problem is not lack of interest, it is pace. Families write at night, on weekends, during a work break. Every hour without a reply costs a visit that did not happen. And once enrollment is over, the same questions keep the team busy all year.

This guide explains how conversational AI helps in both moments, enrollment and routine, and how to organize the enrollment funnel without taking the front office out of the process. It follows what Tolky describes on the Education segment page.

Updated in November 2026.

What is conversational AI for education?

Conversational AI for education is an artificial intelligence agent that talks to families and candidates on WhatsApp, answers with the information the institution registered, logs interest in the CRM and hands off to the front office when a person is needed. It covers the first line of support and hands over the rest with history.

It differs from a menu chatbot by understanding the question the way the family writes it, including audio, with several questions mixed together. For the full difference, see Conversational AI is not a chatbot.

Why does student enrollment need fast replies?

Enrollment needs fast replies because families usually consult several schools at once, and the first to answer well wins the conversation. Three pressures repeat:

  • Enrollment peaks. During the enrollment season, messages multiply and the front office cannot keep up.
  • Repeated questions. Tuition, calendar, supplies and documents: the same questions keep the team busy all year.
  • Prospects without follow-up. Families that ask for information and do not hear back in time choose another school.

This pattern is the same for any operation that lives on contacts received through WhatsApp. See the cost in lead loss on WhatsApp.

What are the four uses of AI in a school?

AI helps on four fronts: immediate answers, enrollment capture in the CRM, community communication and voice. Each has a clear boundary with the front office.

Answers right away, at any hour

For routine questions, the AI answers from the institution's knowledge base and transfers to the front office when it does not have the information. Calendar, enrollment documents, hours, supplies and tuition policy go into the base. When the question is specific to a student, the conversation goes to the front office. To build the base, see how to train the agent with a knowledge base.

Enrollment capture in the CRM

For capture, each interested family becomes a contact with history and a stage in the enrollment funnel. The AI asks what the family is looking for, such as grade or course and campus, logs it in the CRM and hands the admissions team only what is ready for a visit or a conversation. The team picks up already knowing what was said. See the CRM solution and the idea of lead qualification, which also applies to admissions.

Community communication

Text campaigns carry notices and invitations to families, with a conversation when someone replies. They work on WhatsApp and Microsoft Teams, with a Meta-approved template and families' opt-in. Sending outside the 24-hour window follows the rules in 24-hour window and templates. See the text campaigns solution.

Voice for those who prefer to call

The AI answers calls and hands off to the team with the context of the conversation. Telephony depends on activation. To understand the voice layer, see voice AI and semantic IVR.

How do you organize the enrollment funnel in the CRM?

Organize the funnel by stages the team already uses, with the AI moving the contact only as far as it has information. A simple, illustrative design to adapt:

StageWhat happensWho leads
First contactFamily asks on WhatsAppAI
Interest loggedGrade, campus and contact in the CRMAI
Visit or conversationScheduling and welcomeAdmissions team
Proposal and documentsEnrollment, paperwork and termsFront office
EnrolledConfirmation and welcomeFront office

Scheduling a visit is led by the team: the AI records the family's preference and forwards it, without booking a time on its own. The stages and names are an example, not a required standard. The institution defines the funnel that makes sense for it.

What does the AI decide and what stays with the front office?

The AI does not decide scholarships, discounts, enrollment acceptance or policy exceptions. It answers what is in the base and forwards the rest.

SituationConversational AIFront office or admissions
Question about calendar or documentsAnswers from the baseCases outside the rules
Family asking about a courseAnswers, logs in the CRMVisit and proposal
Discount, scholarship or special termsLogs and transfersAnalyzes and decides
Question about a specific studentTransfers with summaryHandles
ComplaintTransfers with summaryResolves

Data about children and teenagers calls for extra care under Brazil's data protection law (LGPD). For the term, see LGPD in customer service and, for the obligations in your case, talk to your legal team. The practical rule is to ask only for what is needed and avoid collecting sensitive data in the chat.

How do you start before the next enrollment season?

The best time to start is before the peak, with time to read conversations and adjust. An illustrative four-week plan usually covers the initial scope.

  1. Week 1: the base. List the 20 most frequent questions to the front office and write the official answers: calendar, documents, hours, supplies and tuition policy.
  2. Week 2: the limits. Define what the AI never decides (seats, scholarships, discounts, exceptions) and the transfer triggers, such as a request for a person or a complaint.
  3. Week 3: the funnel. Agree with the admissions team on the funnel stages and who owns each one.
  4. Week 4: the test. Run with a small group, read the conversations and adjust the base. Only then expand to the whole enrollment season.

For the rules of transition between AI and person, see AI to human handoff rules. To follow what the AI answers, see AI governance in customer service.

In networks with several campuses, it is worth standardizing the core answers and leaving in each campus's base only what changes, such as address, hours and contacts. That way the family gets the same correct answer no matter which campus it reaches.

Which mistakes should you avoid?

The most common mistakes are leaving the base outdated and promising what the school does not decide over WhatsApp.

  • Old calendar and prices in the base. The AI answers what is registered. Review before each enrollment period.
  • Promising a seat or a discount. Define that the AI does not confirm seats, scholarships or special terms.
  • Campaign without opt-in. Notices only for people who authorized contact, with an approved template.
  • Funnel without an owner. Each stage needs someone responsible on the team, or the contact stalls.
  • Nobody reads the conversations. Reserve a weekly slot to review a sample and adjust.

How do you measure whether it is working?

Measure what happens to the family and what the team no longer does by hand. To start, track first response time, the share of conversations resolved without transfer, how many contacts reach the funnel logged and satisfaction by NPS at the end. Tolky uses NPS, not CSAT. Compare with your own previous period, and treat any number from other schools as illustrative.

A useful reading is to separate enrollment from routine. In enrollment, look at how many families reach the CRM with grade, campus and contact filled in and how much time passes between the first message and the team's reply. In routine, look at which questions still go to the front office and turn the most frequent into new answers in the base. That cycle, read, adjust and expand, is what makes support improve with every enrollment period.

Where does Tolky fit?

Tolky is a conversational CRM with AI agents on WhatsApp (official API), webchat and voice, with a history for each contact and handoff to the team with context. Rede Jesuíta de Educação is among the organizations that use Tolky, and the institution appears as a public logo on the site. We do not publish customer result numbers.

Distribution by queues and teams makes it possible to serve networks with several campuses, with centralized history. The AI introduces itself as an AI assistant. Tolky does not promise fully automated support: the design is AI with humans and a handoff with context. To see the support side, go to the WhatsApp solution.

Frequently asked questions

Does the AI answer questions about tuition and calendar?

Yes, based on what the institution registers in the knowledge base. When the question is specific to a student, the conversation goes to the front office.

Can I track enrollments?

Yes. Each prospect enters the CRM with history, and the team sees each family's stage in the funnel.

Does it work for networks with several campuses?

Yes. Conversations can be distributed by queues and teams, and the history stays centralized.

Can I notify families on WhatsApp?

Yes, with campaigns using Meta-approved templates and families' opt-in. Text sending works on WhatsApp and Microsoft Teams.

Does the AI book visits?

The AI records the family's preference and forwards it to the team, which books the visit. It does not promise a time on its own.

Does the AI replace the front office?

No. The AI handles the repetitive volume, and the front office decides scholarships, enrollments and exceptions, with the full history.


Conclusion

In education, conversational AI helps where response time decides: the family comparing schools and the routine that repeats all year. The gain comes from answering right away, logging each prospect and handing off to the front office with context.

Let's talk? See the Education segment, see the plans or talk to the team →

Share

Tags

  • conversational AI for education
  • student enrollment
  • chatbot for schools
  • WhatsApp for schools
  • enrollment funnel
  • #conversational CRM
  • family support
  • #AI agent

About the author

Marlos Carmo

Founder of Tolky

Marlos Carmo is an AI entrepreneur and founder of Tolky, the conversational-era infrastructure and AI CRM that unifies intelligent service, multi-channel support (such as WhatsApp and voice), live CRM, and operational intelligence in a single ecosystem. He is a finalist for the SXSW Innovation Awards and a member of Francesco's Economy, a global network of young entrepreneurs focused on innovation and social impact. He works connecting Artificial Intelligence and digital transformation in projects for large organizations.

Next step

Want to bring this to your operation?

Tati, Tolky's AI assistant, understands your scenario in a few questions, suggests the best path and connects you with someone from the team, with full context.

Read also