Tutoring Facebook Ads: Parent Messaging and Lead Forms With AI

A tutor can run Facebook ads on $150 to $300 a month by writing for the parent, not the child, offering one clear first step such as a free 15-minute chat, and using a short lead form. AI drafts three ad variants and the follow-up message; you check the claims, set a daily budget and answer every form quickly.

Small budgets punish vague ads. At $5 to $10 a day, you have only a few hundred clicks a month to learn from, so the offer, the first line and the form must be right before you spend. The workflow below puts the thinking up front, uses AI where it is quick and good at it, and leaves the decisions that can cost you money, or a complaint, with you.

Follow me on Instagram@sagnikteaches

Decide the offer before you write a word

An ad for "tutoring" competes with every other tutor. An ad for "Year 9 maths confidence: a free 15-minute chat and a first lesson that fixes the topic your child dreads most" competes with almost nobody. Fill in four lines, then give them to the assistant.

Connect on LinkedInSagnik Bhattacharya

• Who the parent is: the parent of a 13-14 year old who is falling behind in maths and avoiding homework.
• What they fear: a lost year, a row at home each evening, paying for a tutor who does not stick.
• What you offer first: a free 15-minute call, then a trial lesson at a reduced price (set whatever you can honestly afford).
• Why you: a specific, true detail, such as a teaching background, the topics you specialise in, or how many students you have taught, as long as it is accurate.

The price and package behind the offer come from your rates work; the tutoring rates and packages tutorial shows how to set them, and the trial-lesson price is a deliberate part of it, not an afterthought.

Subscribe on YouTube@sagnikteaches

Prompt for audience notes and three ad variants

Ask for variety on purpose. Three ads that lead with different motives tell you which worry parents respond to. A prompt that works:

I am a private maths tutor for ages 11-16, online and at home. I want Facebook ads aimed at parents.
Offer: a free 15-minute chat, then a trial lesson at $25.
Parent: has a child aged 13-14 who is falling behind and avoids homework.
Write three ad variants, each with a primary text of 50-150 characters that carries the whole
offer on its own, a headline of 27 characters or fewer, and one optional follow-on sentence.
Variant 1 leads with the evenings-at-home worry. Variant 2 leads with confidence and small wins.
Variant 3 leads with exam preparation and a clear plan.
Rules: write to the parent, never to the child. No guarantees of grades or marks. No claims I have
not given you. No mention of other tutors or schools. Plain English, no hype words.

Those lengths are not arbitrary. Meta's ad guide for a Facebook Feed image ad recommends 50 to 150 characters of primary text and a 27-character headline. Longer text still runs, but a phone shows only the opening lines before "See more", so the first sentence has to sell the free chat by itself. Models ignore length limits surprisingly often, so expect to trim. Here is the sort of output you will get for variant 1 (illustrative), with the fixes underneath:

Is homework turning into a nightly argument? Many parents of 13-14 year olds tell us maths is where it starts. In a free 15-minute chat, we'll find the topic that's holding your child back and show you what a first lesson looks like. Then a trial lesson for $25, online or at home. Book a chat today.

Four fixes. "Many parents tell us" is a vague claim about other people; cut it unless you can say who. "We'll" suggests a team, so keep "I'll" if you are on your own. And "Book a chat today" pushes for urgency you have not earned; use "Choose a time that suits you" and let the form do the rest. Finally, it is far longer than asked: at about 300 characters, the offer sits below the "See more" cut. A trimmed version that does the job in about 130 characters: "Homework rows every evening? A free 15-minute chat finds the maths topic behind them. Then a $25 trial lesson, online or at home." After those edits, the ad reads as one tutor speaking to one parent, which is what you are.

Advantage+ defaults: what to leave on and what to check

Meta's Advantage+ leads page says every campaign with the leads objective starts as "Advantage+ On": one campaign budget that Meta allocates across your ad sets, Advantage+ audience, and all placements, with the system choosing who sees each ad. For a tutor, that is mostly helpful at a small budget, because a narrow hand-built audience often runs out of people. Narrowing the settings too far switches the campaign to "Advantage+ Off", which is allowed; it only means you are steering more of the delivery yourself. Three settings deserve a manual look every time you set up a campaign:

1. Location. In Advantage+ audience, location is a control rather than a suggestion, so Meta stays inside it. Set where you can teach in person, or where you can practically work online, so a widened audience cannot send enquiries from places you cannot serve.
2. Age. Meta's help page on creating lead ads states that you can't target lead ads to people under 18, so a tutoring lead ad reaches adults whatever you choose. To lean towards parents of 11-16 year olds rather than 18-year-olds, either give an age range such as 30 to 55 as a suggestion, or raise the minimum age, one of the hard controls under "Show more options" beside Locations. Expect a raised minimum age to flip the campaign to Advantage+ Off, because the leads page lists "Ages 18 and up" among the conditions for staying On.
3. Audience suggestions. Describe the parent (an age range, interests in education or parenting) as a suggestion. Meta's audience page says it shows ads to people matching the suggestion first and to other people when that is likely to improve results, so read the suggestion as a starting point, not a fence.

The Advantage+ leads tutorial for local services covers the settings in more depth. Whatever the screen calls them this month, keep the principle: let the system optimise delivery, but you decide who you can serve and what you can claim.

The lead form: short enough to finish on a phone

A lead ad opens an instant form without the parent leaving Facebook or Instagram, and Meta may prefill the name, email address and phone number the parent has already given it. Every extra question loses people, and the form is also where you take on legal duties about data. Aim for five fields at most:

1. Parent's full name (Meta collects it by default and may prefill it)
2. Mobile number or email (may be prefilled; choose the one you will actually use first)
3. Child's year or age group, as a multiple-choice question
4. Subject, as a multiple-choice question
5. Best time for a 15-minute call: morning, after school, evening

Add a short privacy line and link to your privacy notice (the form builder has an "Add a privacy policy" box for the link), which must say what you do with the details, how long you keep them and who sees them. Then pick the form type. Meta offers three: More volume (the default, built for quick submission on a phone), Higher intent and Rich creative. Higher intent adds a line under the contact fields saying the business may follow up, plus a review screen where the parent confirms their details before sending. At a small budget I would pick Higher intent, because ten careful enquiries beat thirty accidental taps. One catch from Meta's help page: higher intent forms are delivered only to Facebook Feed and Instagram feed on mobile devices, never on desktop, so the ad will not show in the other placements.

Custom multiple-choice questions, like the year-group and subject fields above, need the form to be built in Ads Manager. Meta's help pages say a form made in Meta Business Suite is limited to a simple contact form, and the quick flow from your Page offers only short-answer questions.

Put a confirmation screen after the submit button: "Thanks, [first name]. I'll reply today with a time for your chat." That sentence is a promise, so only write it if you can keep it. Ask the assistant for a form introduction (two sentences) and the confirmation text; write the privacy line yourself or with your solicitor, since it is the part that counts as a legal statement.

The five-minute follow-up

A form submission is the start of a conversation, not a lead. Parents enquire with several tutors in an evening and tend to go with the first who replies sensibly. Send a short message within minutes and a second one the next day if they are silent. Starter wording to adapt (illustrative):

Hi [first name], thanks for asking about maths help for your child in Year 9. I'm [your name]. Would [day] at [time] or [day] at [time] suit for a free 15-minute chat? I'll ask what your child finds hardest and tell you honestly whether I'm the right fit. If you'd rather email, just reply here.

Notice the last promise: an honest fit check. It costs nothing and it reduces the pressure a parent feels. The full enquiry-to-trial sequence, with the second and third follow-ups, is in the tutoring lead generation tutorial, and it works well alongside the ads here.

Getting the answers into a place you will see them is the job for a small automation, such as the one in the Facebook lead ads to Google Sheets walkthrough. Without it, open Leads Center in Meta Business Suite at least twice a day (you can also download submissions as a CSV from Ads Manager), because a form that sits unread for two days is the same as no form.

The monthly budget, in numbers

These figures are illustrative; your own cost per enquiry may be higher or lower, and it should be measured, not assumed. The point of the table is to show how the money divides, and what the exit rule is.

Now a quick sum for deciding whether the test worked. Suppose a month's $270 produces 18 form enquiries, 9 booked chats and 4 trial lessons, and 2 of those families sign up for regular lessons. Cost per enquiry is $270 ÷ 18 = $15; per booked chat, $30; per trial, $67.50; per new regular family, $135. If one family brings in, say, $40 a lesson for 20 lessons ($800 of revenue), the test paid for itself several times over. If instead you get 18 enquiries and no bookings, the fault is probably in the follow-up or the offer, not the budget, and doubling spend would only double the waste.

Two budget mechanics catch people out. A daily budget is a target averaged over the week: Meta's help page says it may spend up to 75% over it on a good day, but never more than seven times the daily figure in a calendar week, so a $9 budget can show $15 on a Tuesday and still stay within about $63 for the week. And with one campaign budget, Meta's help page says the money flows in real time to the ad sets with the best opportunities and may not be spent equally, even mostly on one. Put each variant in its own ad set so you can see the split, and do not expect $3 each. Write your stop rule before you start: pause any ad that has spent twice your target cost per enquiry (here $30) with no enquiry. The automation tutorial on ad rules, caps and weekly reports shows how to have the account do that for you.

Variants two and three, and how the test reads

The first variant leans on the evenings-at-home worry. The other two should sound different enough that a result tells you something. Illustrative excerpts after editing:

Variant 2 (confidence): Some students do not hate maths. They just stopped believing they can do it. I teach in small steps, aiming for one small win in the first lesson and building from there. Start with a free 15-minute chat. Headline: Maths without the dread.

Variant 3 (plan): Exams feel closer when nobody has said what to do each week. I'll look at your child's topics, set a weekly plan with past-paper practice, and tell you what we are working on after each lesson. Free 15-minute chat first. Headline: A clear Year 9 maths plan.

Both run past 150 characters, which is acceptable as long as the opening sentence works on its own before the "See more" cut: "Some students do not hate maths" and "Exams feel closer when nobody has said what to do each week" each do.

Now a sample three-week result, again illustrative. Twenty-one days at $9 is $189, and Meta did not split it evenly:

Ad 1 wins on volume, ad 2 is the cheapest per booked chat and produced the most trials, and ad 3 is the one to drop. With so few events, none of this is statistically firm, and the uneven spend means ad 2 never had the same chance as ad 1. The sensible reading is "keep 1 and 2, drop 3". If you want a fair head-to-head next, Meta's A/B test tool (in Ads Manager or the Experiments tool) shows each version to a separate part of the audience so nobody sees both, and Meta recommends giving both versions the same budget. Resist the urge to declare a winner from fourteen enquiries.

Advertising rules for children's services

This is where tutors get caught. The rules differ by place, so this is a list of things to check, not legal advice:

• Who sees it. Lead ads cannot be targeted to anyone under 18, and Meta's page on advertising to teens removes lead optimisation for teen audiences altogether. A children's service is always sold to the adult.
• Personal attributes. Meta's ad standards bar ads that assert or imply a viewer's personal attributes, a list that includes age, disability, physical or mental health and vulnerable financial status. "Does your child have dyslexia?" is the kind of question that risks rejection; "Dyslexia-friendly maths lessons, one to one" states what you offer. Read the current standard, not a summary.
• Outcome claims. "Raise your grade by two levels" is a promise you cannot keep. Keep statements to what you do.
• Images of children. Do not use a real student's photograph or work without a parent's written permission for advertising specifically. Prefer a picture of yourself, a desk or a worksheet. If you use an AI-generated image, label it as an illustration where the rules or good sense call for it, and never present it as a real student.
• Testimonials. Quote only what a parent has agreed in writing that you may publish, and keep the message as they wrote it.
• Data from forms. You are collecting a parent's contact details and a child's age group. Keep to what you need, say how you use it, and delete leads that go nowhere on a schedule.
• Local advertising rules. Some places have rules on advertising to or about children and on how prices and free offers must be described. A short question to a solicitor or trade body covers it.

The same caution applies to the wording of your safeguarding promises on any landing page; the tutor safeguarding checklist lists what you can honestly say about checks.

The page behind the ad, and a launch checklist

Parents will tap your name before they fill in a form, so the Facebook page the ad runs from is part of the pitch. A profile picture of you, a two-line description that matches the ad, your subjects and ages, opening hours for calls, and one or two reviews you have permission to show are enough. A page with no posts and no photo makes a cautious parent hesitate. Ask the assistant for the two-line description and three short posts about how a lesson runs; post them before you switch the ad on.

Before you publish, run down this list:

1. The offer in the ad matches the offer in the form and in the first message, word for word.
2. No outcome promise, no real child's image, and no testimonial without written permission.
3. Daily budget set, a total cap set, and a calendar note to look at it on day 3.
4. A dummy form submission has reached your phone or sheet within a minute.
5. The privacy line and link work on a phone.
6. You have blocked time in the first evening to answer enquiries.

Reading the first three weeks

Look at the account at day 3, day 10 and day 21, and decide nothing in the first three days except whether the ads were approved and the form works. Test it yourself with a dummy submission and check that the alert reaches you. By day 10, pause the variant with the highest cost per form. By day 21, compare the survivors on the number that counts: booked chats, not clicks. If the best ad's cost per booked chat is under what a regular family is worth to you, keep it, write a fresh variant on the same angle and carry on. If none are, change the offer (a different year group, a different subject, a lower trial price) rather than the colour of the image.

A realistic mistake looks like this. A tutor sees a cheap cost per form on a "free worksheet" ad, celebrates, and finds that almost none of the form fillers book a chat because they only wanted the worksheet. The fix is to measure from form to booked chat, not from click to form, and to keep any freebie tied to the first chat.

A weekly summary prompt that saves ten minutes

Every Friday, copy the four numbers per ad (spend, form enquiries, booked chats, trials) into a table and ask the assistant to read them for you. A prompt and a plausible reply:

Here are this week's numbers for three tutoring ads (spend, enquiries, booked chats, trial lessons): [table].
Our target is $30 per booked chat. Tell me, in five lines, which ad to pause, which to keep,
and what one change to test next week. Say how confident you are given the small numbers.
Do not invent numbers that are not in the table.

Pause ad 3: $42 for one booked chat, above your $30 target, and it is already getting the least spend. Keep ads 1 and 2, at $28 and $21 per booked chat. With fourteen enquiries in total, confidence is low; treat this as a lean, not a result. Next test: a new third variant aimed at Year 11 parents, leading with a revision plan.

The line to watch is the confidence statement. If the assistant ever sounds certain on single-digit numbers, ask it to redo the answer and say what the sample size allows.

What AI should and should not do in this workflow

Use it for first drafts of copy, variations on a good ad, the form introduction, follow-up wording, and a weekly summary of the numbers you paste in. Do not use it to decide your budget, to write claims about yourself that you cannot document, to generate student testimonials, or to create images that look like real children. Everything it writes is yours to check, and the most useful check is reading each ad as a sceptical parent: would I believe this, and would I be happy if the tutor did exactly what it says?

Questions tutors ask before running ads

How much should a tutor spend on Facebook ads to start?

A test of $5 to $10 a day for three to four weeks, so roughly $150 to $300 a month, gives enough clicks to compare two or three ads without risking much. Treat the first month as a measurement of your cost per enquiry and per trial lesson, not as a way to fill your diary.

Should the ad target children?

No, and for a lead ad Meta does not allow it: its help pages say lead ads cannot be targeted to anyone under 18. Write to the parent or carer, who is the buyer, and keep the wording about what you offer rather than about the child. Do not use a child's photograph without a parent's written permission for advertising.

Can I promise better grades in the ad?

Avoid it. Results depend on the student, and platforms and advertising regulators dislike unsupported outcome claims. Describe what the lessons do (weekly practice, feedback, past-paper work) and let a parent's own testimonial, with permission, say the rest.

Do I have to say the ad was written by AI?

AI-drafted wording that you have checked and edited is your copy, and you are responsible for it. If you use AI-generated images, check how Meta labels or requires you to disclose them, because those rules have been changing. Never present a generated image as a real student or real result.

Further reads

• How to Get Tutoring Clients With AI: Profile, Referrals, Flyers — The free routes to clients that sit beside paid ads.
• Roofing Facebook Ads: Copy and Audiences Tested With AI on a Small Budget — A worked ad test from another local service, for comparison.
• Carpet Cleaning Facebook Ads: Write and Test Offers With AI — How another small service tests offers one at a time.
• Pressure Washing Facebook Ads: Before-and-After Creative Made With AI — Creative testing on a small budget, with checks on claims.
• Tutor Scheduling Software: Bookings, No-Shows and AI Reminders — Booking tools that turn a chat into a confirmed slot.
• How to Start a Tutoring Agency: Tutors, Pricing and AI Admin — The margin per lesson hour, tutor and parent terms, safeguarding basics and first-month admin for a new tutoring agency, with AI doing the drafting.
• AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.

Sources: Meta Business Help Center pages on instant form types, lead ads with instant form, creating a lead ad from a Page, advertising to teens, daily budgets, Advantage+ campaign budget and A/B testing; Meta's Advantage+ leads and Advantage+ audience pages; Meta's Facebook Feed ad guide; Meta's advertising standard on privacy violations and personal attributes (all read November 2026).