Tutoring Centre Admin With AI: Enquiries, Timetables, Invoices

A tutoring centre can use AI to draft enquiry replies from a fact sheet of subjects, prices and availability, to propose timetables that a spreadsheet then checks for clashes, and to flag credits, discounts and missed lessons before termly invoices go out. A person approves each step; the time saved comes from never starting from scratch.

Which of the three to tackle first depends on where your hours actually go, and owners are often wrong about that. A centre that feels buried in enquiries may find that the real drain is rescheduling, and fixing the wrong one first means months of small gains. A one-week time log settles it.

Follow me on Instagram@sagnikteaches

A one-week admin log before you change anything

For one ordinary week in term, whoever does admin notes each task and roughly how long it took. An illustrative, filled-in summary for a centre with 160 students:

Connect on LinkedInSagnik Bhattacharya

In this example, enquiries and timetabling together take ten hours a week, and invoices spike at the start of term. That suggests starting with enquiries (quick to set up), then timetabling, and having the invoice process ready before the next term starts. If you haven't yet decided whether any AI subscription is worth paying for, whether AI is worth paying for in a small tutoring business helps with that call.

Subscribe on YouTube@codingliquids

Enquiries: a fact sheet, three questions, and a draft in seconds

Most enquiry replies repeat the same facts. Write them down once in a one-page fact sheet: subjects and levels, group and one-to-one prices, session lengths, term dates, how the free assessment or trial works, the cancellation policy, and which slots currently have space. Update the availability line weekly. A few lines from an illustrative sheet:

• Maths, ages 11 to 16: small groups (up to 4) Mon to Thu 16:00 to 19:00 and Sat 09:00 to 13:00; one-to-one by arrangement.
• Prices: group $38 an hour, one-to-one $55 an hour, billed termly; 10% off the second sibling.
• Assessment: free, 30 minutes, Saturdays or weekday 16:00 slots; parents can stay.
• Cancellations: more than 24 hours' notice earns a credit or a make-up lesson; less notice is charged.
• Space this week (updated Mon 7 Oct): Sat 11:00 maths group (1 place), Wed 17:00 English group (2 places).
• Not offered: coursework marking; specialist support for learning difficulties.

The "not offered" line matters as much as the rest, because it's what stops a draft agreeing to things you don't do.

Then give the assistant the fact sheet with each enquiry and ask for a reply that answers what the parent asked, asks at most three questions you need answered, and offers a trial slot. A realistic enquiry:

Hi, my son is 14 and really struggling with maths, his teacher says he needs support before his exams next summer. Do you do Saturdays? How much is it? He's quite anxious about it.

An illustrative draft, after a light edit by the office manager:

Thanks for getting in touch, and it's good that you're starting well before next summer. Yes, we teach maths on Saturdays, with small-group sessions (up to four students) at 9:30 and 11:00, and one-to-one sessions by arrangement. Small-group sessions are $38 per hour and one-to-one sessions $55, billed termly.

We start with a free 30-minute assessment, which is relaxed and mostly a chat about what he finds hard; lots of students arrive anxious and it helps them meet their tutor first. Could you tell me which topics he finds hardest, whether he prefers working in a group or alone, and whether this Saturday or next suits for the assessment?

Two things make this work. The fact sheet stops the assistant from inventing prices or times; tell it to write "CHECK" if a fact isn't on the sheet. And the anxiety comment gets a human sentence, which the office manager added. For speed targets and how to hand drafts over quickly, see replying to every enquiry in under five minutes.

Some enquiries shouldn't get a drafted reply at all. A parent writes: "My daughter has ADHD and had a really bad year, she missed half of last term. Can you help her catch up in English?" An illustrative first draft answered warmly and added, "Our tutors are experienced in supporting students with ADHD." Nothing on the fact sheet says so, and the sheet's "not offered" line says the opposite about specialist support. Sent as written, that sentence is a promise the centre can't keep, made to a parent who has had a hard year. Add a rule to the prompt: if an enquiry mentions a diagnosis, additional needs, long absences from school, wellbeing or anything about safeguarding, don't draft; label it "owner to reply" and send it to the person who talks to those families.

Timetables: let AI propose, let a spreadsheet catch the clashes

Timetabling is a constraint puzzle: tutors' availability and subjects, room capacity, students' availability and levels, group sizes, and sibling pick-ups. AI is decent at proposing a first arrangement, and unreliable at guaranteeing there are no double bookings. So split the job.

First, write the constraints as a list the assistant can read:

Tutors (subject, available slots):
- T1 maths: Mon-Thu 16:00-19:00, Sat 09:00-13:00
- T2 English: Tue, Thu 16:00-19:00
- T3 maths and physics: Sat 09:00-13:00
Rooms: R1 (4 students), R2 (4), R3 (1-to-1 only)
Students (subject, level, group or 1-to-1, available):
- S014 maths, higher, group, Sat morning only
- S022 maths, foundation, group, Mon or Wed after 17:00
- ...
Rules: max 4 per group; group by level; siblings S031 and S032 at
the same time; no tutor teaches more than 3 hours without a break.
Propose a weekly timetable as a table: day, time, tutor, room,
students. List any student you couldn't place and why.

Then paste the proposed timetable into a sheet with one row per student per session (columns: Day in A, Time in B, Tutor in C, Room in D, Student in E) and add three checks:

Tutor clash    =COUNTIFS(A:A,A2,B:B,B2,C:C,C2)>COUNTIFS(A:A,A2,B:B,B2,C:C,C2,D:D,D2)
Room over cap  =COUNTIFS(A:A,A2,B:B,B2,D:D,D2)>VLOOKUP(D2,Rooms!A:B,2,FALSE)
Student clash  =COUNTIFS(A:A,A2,B:B,B2,E:E,E2)>1

The first flags a tutor booked in two different rooms at the same time; the second, a room over capacity (with room capacities on a sheet called Rooms); the third, a student in two places at once. Any TRUE needs fixing. In an illustrative first run for 160 students, the assistant placed 151, listed nine it couldn't fit, and the checks caught two clashes it had missed: a tutor in R1 and R3 at 17:00 on Wednesday. The nine unplaced students are the ones worth a person's attention, and the sheet tells you exactly who they are.

For mid-term changes, the same approach works in miniature: "S022 can no longer do Mondays; which existing groups at her level have space on Tuesday or Wednesday?"

An illustrative answer to that question: "Tuesday 17:00 with T1 in R2 has 3 students, so 1 place." Check it before offering it. The Tuesday group is higher level and S022 is foundation, so the suggestion breaks the rule about grouping by level; the constraint list the assistant was given for this quick question didn't include levels. Always paste the same constraint list, even for a one-student change, and run the new row through the clash checks before emailing the parent. If group sizes and attendance data are part of the question, using booking data to fix a class timetable goes further.

Tutor absence is the other mid-term job, and it's where a drafted message saves the most phone calls. When T2 calls in sick on a Thursday, give the assistant the day's sessions for T2, the free tutors and rooms, and the rule "offer a cover tutor for group sessions, a reschedule for one-to-ones". It returns a list of affected students and a short message per family: "[first name]'s English group at 17:00 today will be taught by another of our English tutors, same room and time. If you'd rather reschedule, reply and we'll offer a make-up lesson." The office manager checks the cover tutor really teaches that level, then sends.

Invoices: built from attendance, with the exceptions flagged

Termly invoices are simple until the exceptions arrive: siblings, make-up lessons, notice given for absences, mid-term starters, and a family paying in instalments. The reliable pattern is to let your billing tool produce invoices from lesson records and use AI to review the exceptions.

If you use a tutoring management system, invoicing from lessons is usually built in. Teachworks, for example, can generate invoices individually or in bulk for lessons, packages or flat fees, with an Invoice Autopilot add-on to automate them; TutorCruncher has an Auto Invoice setting that raises and sends an invoice as soon as a lesson is marked complete, one invoice per lesson, which suits pay-as-you-go families better than termly billing. If you use accounting software with repeating invoices instead, the tricky part is adjusting each term for exceptions, which is where an assistant helps.

Before the invoices go out, give the assistant your policy and the term's attendance and changes, and ask it to list every family whose invoice should differ from the standard amount, and why. It should not calculate the totals; your billing tool or a sheet does that. An illustrative output:

Family F-061: two siblings in group maths; sibling discount (10%) applies to the second child.
Family F-078: missed 2 sessions with more than 24 hours' notice; policy gives a credit of 2 sessions or 2 make-up lessons. No make-ups recorded, so credit 2 sessions.
Family F-090: started in week 4; invoice 9 of 12 weeks.
Family F-102: missed 1 session with no notice; policy says charge as normal. No change.

The office manager checks each line against the records and applies them in the billing tool, doing any sum on a calculator rather than trusting a total. For F-090, a one-hour weekly group at $38 over 9 of 12 weeks is $342 instead of $456; for F-078, two one-hour group sessions credited is $76 off the term. If the billing tool shows anything else, find out why before sending. A realistic mistake this catches: in one illustrative term, a sibling discount had been added manually to a family's repeating invoice and then applied again by a new discount rule, charging them 20% less than intended. Listing exceptions every term surfaces that kind of double adjustment.

Unpaid invoices then get a polite sequence: a reminder a few days after the due date, a second a week later, then a phone call. Chasing late payments with AI reminders that sound human has wording you can adapt for parents.

Instalment plans need to reach the reminder sequence, or it undoes the goodwill. Picture a family who agreed to pay the term in three parts: the first instalment is paid, and the automatic reminder a few days after the due date tells them the "remaining balance of $912 is overdue". It isn't; it's on the plan. Put "instalment plan: dates and amounts" on the exceptions list each term, and set those invoices to remind only on each instalment date, for that instalment's amount.

Joining the three together: enquiry to first invoice

The real saving comes when each step hands data to the next rather than being retyped:

1. The enquiry reply offers an assessment slot; the parent books it through an online booking link (Teachworks, for instance, supports booking from your website).
2. After the assessment, the tutor records subject, level and preferred times in the student record.
3. The weekly timetable prompt picks the new student up from the record, and the clash checks run.
4. The confirmed slot becomes the lesson record that the invoice is built from.

Nothing here needs complex automation. A shared student sheet or your tutoring system, with a consistent student number used everywhere, is enough.

An illustrative centre with 160 students, 11 tutors and 5 rooms

Using the log above as the starting point, here is what a realistic first term might look like after setup:

That's about 10 hours a week back, from roughly 17 to 7. Setup takes about two days spread over three weeks: the fact sheet, the constraint list, the clash-check sheet and the invoice exception prompt. The running cost is an assistant plan for the office, around $20 to $25 a month per seat; if two people need access, a Business or Team plan requires at least two seats. These figures are illustrative, and the log you keep after the changes is the real test.

What goes wrong, and how you'll notice

• The fact sheet goes stale. A price rise or new term dates don't reach the sheet, and drafts quote old figures. Put a "last updated" date at the top and a weekly calendar reminder.
• Constraints left out. The assistant can't know a tutor won't teach two back-to-back higher groups unless it's in the list. When a proposed timetable looks wrong, add the rule rather than fixing it silently.
• AI arithmetic. Asked to total a term's invoice, assistants make small errors. Totals come from your billing tool.
• Children's data in the wrong place. Use student numbers in timetabling prompts, and a business plan for anything with names.

Once the admin is under control, the next job most centres hand to AI is reporting to parents; how tutoring centres use AI to write parent progress reports picks up from there.

Admin questions from tutoring centre owners

Do we need tutoring management software, or can we manage with AI and spreadsheets?

Under about 60 students, a shared spreadsheet, an accounting package with repeating invoices and a chat assistant usually cope. Past that, the number of reschedules, make-up lessons and credits starts to break spreadsheets, and a tutoring system that links lessons to invoices saves more time than any prompt. Trial one with a single term's data before moving everyone.

Is it safe to put children's names into an AI tool?

Only in a tool your centre has approved, on a business plan that doesn't train on your content by default, and only what the task needs. For timetabling, initials or student numbers work as well as names. Tell parents in your privacy notice which tools process their child's information.

Can a chatbot on our website answer parent enquiries?

It can answer factual questions from your fact sheet, such as subjects, prices, term dates and how trials work, and collect enquiry details. Anything about a specific child's needs, safeguarding or complaints should go to a person. Start with drafted email replies, and add a chatbot once you know which questions come up most.

Further reads

• How Tutors Can Plan a Week of Lessons With AI in an Hour — Save tutor time on lesson planning as well as admin.
• How Tutors Should Check AI-Made Worksheets Before Using Them — Checks before any AI-made worksheet reaches a student.
• Answering Pupil and Parent Enquiries With AI at a Driving School — Another lesson-based business answering parent enquiries.
• Is It Safe to Use AI With Children's Data in a Nursery? — Handling children's data with AI tools safely.
• How to Build the FAQ Your AI Chatbot Needs Before Launch — Build the fact sheet a chatbot or assistant needs.
• AI Appointment Scheduling for Small Businesses: Tools and Setup — Scheduling tools for trial lessons and assessments.
• AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.

Sources: Teachworks tutoring management software pages (invoicing, reminders, online booking); TutorCruncher help pages (Auto Invoice setting).