How can a small class turn a trial enquiry into a confirmed booking?

Check the class fit before offering a slot

Publish the subject, intended participants, duration, venue, trial price or conditions, what to bring and cancellation rule. Ask one relevant question such as adult versus family, prior experience or preferred day. Do not infer a person's ability from a sentence. Age or safety boundaries must come from the teacher's approved rules.

Separate Saturday preference from the class, date, individual trial slot and group-class seat. Native kakinang booking shows configured service and staff time slots and lets a customer preview and confirm one slot; it has no participant-count or seat-capacity field. A class seat needs staff verification or a separately checked existing enrolment system. Do not treat an open staff slot as a remaining group seat. Without a schedule, AI can only record a preference awaiting staff review. After a new class or teacher change, check the public page, timetable and knowledge together. Google's guidance allows eligible businesses to add booking links, but an external listing still needs owner review; it does not imply automatic synchronization.

For example, two parents ask about a family pottery class, with children aged five and ten. If the published class minimum is eight, respond to each against that rule. Suggest an appropriate alternative or teacher review for the younger child. A model should not make its own judgement just to fill a place.

Collect only the data needed for a trial: name, suitable contact route, participant count and conditions relevant to teaching. Do not demand full family details before a place is even available. For minors, the school sets its parent-contact and consent process and routes sensitive cases to an authorized person. State date and time zone clearly, especially for an online class across countries; 'Saturday afternoon' may mean different times.

Any displayed group seats must come from the school's maintained roster or separately verified enrolment system, never native kakinang capacity. Avoid unsupported last-place urgency. If the school runs a waitlist, identify it as staff-managed with a responsible notifier; it is not a native booked seat. When cancelling a class, care for already confirmed students before leaving old advertisements pointing to it.

Use preview and confirmation for a real commitment

For an individual trial or consultation using native booking, review the service, date, staff time, venue and published cancellation rule before confirmation. The native preview does not carry participant count or group-seat capacity. It does not hold a class seat. For group classes, staff must check places and conditions or use a separately verified enrolment system; AI must not claim that the last seat is protected by native booking. On success provide formal status details; on failure say no booking was created and offer another time or a staff route.

If the school wants only to collect interest, use a lead instead: preferred class, date range and consented contact method. A lead must not trigger a 'your class is booked' message. Staff inspect the queue, check capacity and invite the customer to formal confirmation. The assistant prevents a missed enquiry; it cannot add a seat.

A booking link is useful only if it leads to the same class and date described on the page. Walk the mobile path from class description through availability, preview, confirmation and completion. Inconsistent labels between the website and the booking page can cause the visitor to choose the wrong class.

Handle status, cancellation and exceptions

Public booking status and cancellation require the existing ownership or email verification. A stranger should not cancel another person's place by knowing a name. For rescheduling, read the current status and policy, then use formal cancellation and a new booking or send the case to staff. Duplicate confirmation should not create two places, but operators still need to inspect duplicate requests and notifications.

Prepare answers for a full class, a teacher's unexpected absence and an individual suitability concern. A full class gets alternatives verified by staff; an absence requires schedule and page updates plus care for existing bookings; suitability goes to the teacher. Email and human handoff require configured recipients, SMTP and actual staffing before anyone can claim a teacher was alerted immediately.

Weekly, count trial enquiries that became confirmed bookings, preferences still awaiting reply and failures caused by conflicting schedules or rules. Test a class found full during staff review, a mistyped date, an old link after cancellation and a cancel-then-rebook flow. Fast AI replies alone cannot show whether a student really has a place.

Write the class rule for visitors, not only as an internal code: 'For beginners aged eight and above, six places, arrive ten minutes early, materials included.' Make clear whether make-up rules differ between trials and regular enrolment. AI can explain approved wording, but unusual support needs, age exceptions or extra places require teacher review. An apparently open slot does not authorize a policy exception.

Capacity can depend on resources as well as chairs. One instructor cannot teach two rooms at once, and a kiln or materials may be shared. Native booking configures service and staff time, not classroom resource or group-seat capacity. Check teachers, room, materials and remaining seats in the school roster or a separately verified existing enrolment system. If these limits cannot be checked, take an enquiry for human review rather than let AI promise a seat.

Customer messages must distinguish 'we received your request' from 'your booking is confirmed'. The former states a follow-up expectation; the latter lists class, time, venue and status or cancellation route. Missing email alone does not prove a booking does not exist; verify the formal status and back-office record. Review weekend questions, incomplete confirmations and places reopened after cancellations each week to find missed follow-up or unclear rules.