Can You Sell Booking and Scheduling Data for AI Training?

Partly. The scheduling records of a business that sends people out to do skilled work (jobs, dispatch, technicians' notes, the calendar that ran them) have named buyers: five buyers of company records name scheduling, dispatch or field-service records, and two of them list ServiceTitan among the sources they buy. A list of customer bookings on its own (who booked which slot at a salon, a studio or a clinic front desk) has no buyer that names it, and no public price. Both sell, where they sell, as part of the company's whole working record.

The booking system is rarely what stands in the way. The terms of the eight booking and field-service systems we read leave the records with the business, and none of the clauses we read bars a business from licensing its own copy, with one exception for franchisees. What decides most sales is what the records show about the work, and the people whose names, numbers and appointments are in them.

This guide is for founders and managers of businesses that run on bookings or jobs: trades and home-service firms (HVAC, plumbing, electrical, roofing, cleaning), salons, spas and fitness studios, clinics and therapists, tutors and schools, and any business whose day is a calendar of appointments. It covers which buyers want scheduling records and why, what your booking system's terms say, what the records are worth, the customers and staff in them, how to take them out and prepare them, and two worked examples. It sits under what AI labs buy, beside the guides to CRM data, support tickets and financial records.

Contents

• What booking and scheduling data is
• Which buyers name it
• Why the job record matters more than the booking
• What your booking system's terms say
• What booking and scheduling data is worth
• The people in the bookings
• Taking the records out and preparing them
• Two worked examples
• Questions people ask
• What this means for you

What booking and scheduling data is

"Booking data" covers three different records, and they do not sell alike. The first step is to name which of them your business holds, because a buyer reads each one differently.

Appointment bookings are the record of a service business that sells time: a salon, a spa, a fitness or yoga studio, a physiotherapist, a tutor. The system (Fresha, Mindbody, Square Appointments, Acuity, SimplyBook.me, Calendly and the like) holds who booked, which service, with whom, when, what was paid, the no-shows and cancellations, and often an intake form and a few notes. Reservations are the same thing for a restaurant or a venue, usually held by the reservation platform. Jobs and dispatch are the record of a business that sends people out: a call or a request becomes a booked job, a technician is dispatched, the work is done and written up, an estimate becomes an invoice and a payment. Field-service systems such as ServiceTitan, Housecall Pro and Jobber hold that whole chain.

The difference matters because buyers of company records pay for how work is done. A booking says that something happened at a time. A job record says what the problem was, what the technician found, what was done about it, what it cost and how it ended. The second is evidence of a skill, and that is what buyers say they want.

Many businesses hold more than one. A home-service company books appointments and dispatches jobs in the same system; a clinic holds appointments beside clinical notes; a studio holds bookings beside its class programme and its staff rota. The record that sells is usually the one that joins the booking to the work.

Which buyers name it

Buyers of company records publish what they want on their own pages, and we read them for booking, appointment, scheduling, dispatch and field-service records on 1 November 2026. The result splits cleanly: scheduling, dispatch and field-service systems are named by five buyers, customer bookings by none.

Nyne names it outright. One of its three categories of company data is scheduling and delivery: jobs, dispatch, calendars, Asana or Jira, and the project and service records that show work moving start to finish. Handshake AI and Mode (BizBuyData) both list ServiceTitan, the field-service platform, among the sources they buy, beside Slack, Gmail and QuickBooks. Scale AI lists field services (HVAC, plumbing, electrical) among its industries of interest. Mercor's target industries name dispatch logs for logistics companies. HUMXN Research Labs, which buys records from trades businesses, lists for home services service calls, quotes, job notes, technician photos and dispatch.

Nobody we read names the appointment book of a salon, a studio or a clinic. Mercor publishes listings by kind of company, and none of them is for a business that runs on bookings: there is a listing for medical practices, for construction and for call centres, none for a salon, a studio, a restaurant or a home-service firm. That is not a rule that such records cannot sell. It means a buyer will judge them as part of the company's whole record, by what the rest of that record shows.

The size lines matter as much as the names. Most buyers that publish one start at 20 or more full-time employees, and the Scale AI programme prefers 40 or more. A two-chair salon or a sole trader falls under every floor we have read, and the guide to buyer programmes sets the floors side by side.

Why the job record matters more than the booking

What buyers say they want is the record of skilled work, step by step. Pewter puts it plainly: records that show skilled work step by step are in highest demand, and it names trades and professional services among them. The model a buyer trains learns what a person in that job does next, so the value is in the steps, not the time slot.

A field-service system holds those steps joined together. Polyshares explains why it values ServiceTitan data: the system links a call, a booked job, the technician's work, the invoice and the payment in one chain, which shows which diagnoses led to which repairs and what the customer paid. That chain is a worked example of a trade, thousands of times over, with the outcome attached.

The same buyers want the work itself, recorded. HUMXN Research Labs runs a programme that pays for plumbing, electrical and HVAC appointments in three US cities so that the job can be recorded on a wearable camera, and it has announced $50 million to acquire and license real-world information from skilled trades and manufacturing. Footage of that kind is covered in the guide to robot and first-person data; the point here is that the trades are wanted for their work, and the scheduling record is how that work is found, ordered and dated.

An appointment book, by contrast, mostly shows demand: what was booked, when, and who did not turn up. That has uses inside the business (staffing, pricing, reminders), but a buyer of company records cannot learn a skill from it. What can lift an appointment-led business's record is what sits beside the bookings:

• Notes on the work: a physiotherapist's treatment notes, a colour formula at a salon, a trainer's programme, where the business holds the rights and the law allows it.
• Conversations: the messages and calls in which customers asked, were advised and booked, joined to the booking they produced (the WhatsApp guide and the sales calls guide explain why the outcome joined to the conversation is what buyers pay for).
• Procedures: how the business trains staff, handles complaints and runs its day, the documents Troveo calls hard to find anywhere else and written by people who do the job.

So the honest line is this. A trades or field-service business has named buyers for its jobs, dispatch and calendar history. A business whose records are customer bookings has none that names them, and offers them, if at all, as one part of a fuller record.

What your booking system's terms say

Your booking or field-service system holds the records, so its terms are the first document to read. We read the terms of eight systems on 1 November 2026, each from the vendor's own page. The pattern is the same across most of them: the business keeps its records, the vendor keeps a broad licence, and several vendors own the aggregate and train their own AI.

Ownership: the records are the business's

Six of the eight say in so many words that the business keeps its data, SimplyBook.me makes the business the controller and itself the processor, and Fresha has the business warrant that it owns what it puts in. None of the clauses we read says the business may not license its own copy to someone else. That is a statement about the clauses read, not about every document a vendor signs: an order form, an enterprise agreement or a data processing addendum can say more, so read yours.

The vendor's licence and its aggregate

Several vendors hold a licence of their own that a buyer will want to hear about. Mindbody's and Square's are perpetual, irrevocable and sublicensable; Square's covers developing, training, testing, evaluating and improving models; Fresha may sublicense and offer the content to third parties. A licence to the vendor does not stop the business licensing its own copy, but it does mean the business cannot promise a buyer that nobody else holds or uses the records.

Three vendors also own the anonymised or aggregated layer built from their customers' data: Mindbody, Housecall Pro and ServiceTitan. A business can license its own records, never the vendor's benchmarks, dashboards or industry statistics.

The vendor's own AI, and what it means for exclusivity

Housecall Pro's terms record the user's consent to its data training Housecall Pro's AI models, and say data used that way may stay in the model. Mindbody may train its AI on de-identified or aggregated data without compensation, and takes a waiver of claims. ServiceTitan goes furthest the other way: it will not use Customer Data to train AI models for the benefit of any third party.

This matters most when a buyer asks for exclusivity, or for data no model has seen. A business on Housecall Pro or Square cannot honestly promise either for the years its records sat there. The guide to exclusive and non-exclusive licences explains how to bound an exclusive grant so it promises only what the business controls.

Franchisees: the one outright bar

ServiceTitan's terms hold the only clause we found that stops a business licensing its records at all. For a franchisee, the franchisor, and not the franchisee, owns all Customer Data unless the franchise agreement says otherwise, and the franchisee may not seek a mass export without the franchisor's consent. A franchised trades business starts with its franchise agreement and its franchisor, not with a buyer.

The systems we could not read

We could not read the terms of Jobber, Booksy, Vagaro, Zenoti, OpenTable or Resy when this guide was written: their pages refuse scripted reading or render only in a browser. If your business runs on one of them, read the sections on customer data, licences to the vendor, aggregated data and export before anything else, and keep a copy of the version in force.

What booking and scheduling data is worth

No buyer, listing or disclosed sale we found prices booking, appointment or scheduling records on their own. Every buyer that names them prices the company's whole record, and the scheduling history is one part of it. The figures that apply are therefore whole-company figures, and they apply best to the businesses buyers name: trades, field service and logistics.

Across the sales of company records on record, the median is $72.5K per company a company, and published offers to holders run at a median of $278K per company. The buyers' own ranges are wide: Handshake AI says it has paid $100K to $4M+ per partnership, Mode (BizBuyData) $100K to $5M a company, Nyne commonly $100K or more, and Scale AI $10K to $1M+.

The closest published sales are for the businesses this guide is about:

None of these is a price for scheduling records. Each is a bundle in which jobs or dispatch were one part, sold by a firm large enough to clear the buyers' size lines, and two of them were exclusive, which buyers pay more for. A smaller business can see what one buyer's own estimator says: Nyne's, run for a US company of 10 to 24 staff in business 3 to 7 years whose accounting, customer and scheduling records it would license, shows $170,000 to $299,000. That is an estimate on a buyer's page, not an offer, and the guide to how buyers calculate offers explains how far such numbers move once the records are seen.

For a business that holds only appointment bookings there is no figure to give. What it can do is value the whole record, with the bookings as one system among the others, and see what that record shows. The guide to what company data is worth sets out what raises and lowers a company's price, and the market rates page shows every price behind these figures.

The people in the bookings

Every booking names a person: a customer with a phone number, an address for a home visit, a time they were somewhere, often a note about them. A licence of booking records is a licence of personal data until those people are taken out, and three groups appear in the records.

Customers. Under California's privacy law, licensing personal information to an AI buyer for money is a sale, for a business over the law's thresholds, and a business that sells must first tell consumers and honour their opt-outs. Records that are properly deidentified fall outside those rules, which is why de-identification is the usual route. Bookings for children (a swim school, a tutor, a children's class) need more care: their information cannot be sold without affirmative opt-in. In Canada the privacy regulator expects consent for new purposes and disclosures to third parties that use the data for their own purposes; in the UK a new purpose is weighed under the compatibility test. And in the US, using records in a way a privacy policy promised not to can be an unfair or deceptive practice.

Field-service records hold more than a booking does. A home visit carries an address, and often an access code, a note about who is at home, photos taken inside the house and the customer's own description of a problem. Removing names is not enough when an address or a photo still points to one household. Canada's privacy regulator found, in a case about anonymised data, that taking out names, emails and phone numbers did not meet the standard of no serious possibility of re-identification. The de-identification standard sets out how records are cleaned before they leave.

Health bookings. An appointment with a doctor, a dentist, a physiotherapist or a therapist says that a person received care. For a provider that HIPAA covers, those appointments are protected health information, and HIPAA binds most providers and the vendors that handle their data. Such a provider cannot take money for protected health information without each patient's authorization; data that meets HIPAA's de-identification standard is no longer protected. Businesses HIPAA does not cover are not free of health rules: Washington's law reaches health data held by businesses HIPAA does not cover, such as wellness and fitness services, and selling identifiable consumer health data needs a signed authorization naming the buyer, one per sale. A massage, a weight-loss programme or a booking with a nutritionist can be health data. The HIPAA de-identification guide and the page for medical practices go further.

Staff. A scheduling system is also a record of staff: rotas, hours, which technician went where, their notes and sometimes their location. In California, the privacy regulator states that employees are consumers with rights under the law. Staff details come out of a lot the same way customers' do, and the guide to employee emails and chats covers what an employer may do with its staff's records.

The booking system's own terms point the same way. SimplyBook.me makes the business solely responsible for obtaining a valid consent from its clients, and Fresha separates the business's own clients from clients who booked through Fresha itself, whose data the business may reach only for the term. A lot of Fresha bookings has to be split by whose client each person is. The guide to whether it is legal to sell customer data covers the general rules.

Taking the records out and preparing them

Take the records out through the system's own export, reports or API. Mindbody's terms bar systematically downloading and scraping its content, and Square's bar robots, spiders and scrapers on its systems; neither clause limits what the business does with an export it is given. Take the export while the account is live: ServiceTitan's export after termination is a request it makes commercially reasonable efforts on, fees possible, with data kept no more than 60 days. A business that is closing should export before it cancels, and the guide to who owns a dissolved company's data covers the steps after closing.

Before any scheduling records go into a lot:

1. Name each system and its years. Which booking or field-service system held the records, and from when to when. A business that moved systems has two exports with different fields.
2. Read the terms in force for those years. Look for who owns customer data, licences to the vendor, aggregated data, the vendor's AI training, and anything about franchises. Note what the vendor already holds, so a buyer hears it from you.
3. Join the bookings to the work. Jobs to their notes, photos, estimates and invoices; appointments to the conversations and outcomes around them. The joined record is the one buyers name.
4. Leave out what you cannot stand behind. Health bookings without authorization or de-identification, children's records, clients who belong to the booking platform rather than the business, and anything the privacy notice for those years did not cover.
5. De-identify the people. Names, phone numbers, emails, addresses, access codes, payment details, and faces and house numbers in photos come out; staff names are replaced with stable codes.
6. Describe it in a datasheet: the systems, the years, the number of jobs or bookings, what each record holds, what was removed and what was left out and why.

The guide to preparing data for sale covers the rest of the company's systems, and the rights check asks the questions above in order.

Two worked examples

These examples are illustrative.

An HVAC company. A heating and cooling firm with about 40 staff, independent rather than franchised, has run on ServiceTitan for nine years. Its records hold every call, booked job, dispatch, technician note and photo, estimate, invoice and payment, beside its QuickBooks books, its email and its training manuals. This is the record buyers name: Nyne lists jobs and dispatch, Handshake AI and Mode (BizBuyData) list ServiceTitan, and Scale AI lists field services. The firm clears the size lines of the buyers that publish one. ServiceTitan's terms leave the Customer Data with the firm and commit ServiceTitan not to train models for third parties on it, so the firm can offer exclusivity over its own records, though not over ServiceTitan's aggregate. It exports while its account is live, removes customers' names, numbers, addresses and access codes, blurs faces and house numbers in photos, codes its technicians, and joins each job to its notes and invoice. Offered as one record, it sits among the trades sales on record, such as a roofing company's $450,000 exclusive licence.

A beauty salon. A salon with eight staff has taken bookings on Fresha for six years. Its records are appointments (client, service, stylist, time, price, no-shows) and a few notes on colour formulas. No buyer names appointment books, and eight staff falls under the size line of every buyer that publishes one. Some of its clients booked through Fresha itself, and the salon may reach their data only for the term of its agreement. The honest answer is that the bookings alone are unlikely to sell. If the salon is part of a larger group, or holds richer records (training material, client consultations, a product line's sales), those are what to value first, with the bookings as one supporting system.

Questions people ask

Can I sell my booking data to AI companies? If your bookings are jobs and dispatch in a trades or field-service business, yes: five buyers name such records, as part of the company's whole record. A list of customer appointments on its own has no buyer that names it.

Does my booking software own my data? In the eight systems we read, no: the business keeps its records, and SimplyBook.me holds them as the business's processor. Several vendors keep a perpetual licence, own the aggregated data and train their own AI. A ServiceTitan franchisee's data belongs to the franchisor unless the franchise agreement says otherwise.

Do AI companies buy ServiceTitan data? Handshake AI and Mode (BizBuyData) both list ServiceTitan among the sources they buy, and Polyshares values it because it links a call, a booked job, the work, the invoice and the payment in one chain. They buy it with the rest of the company's record, not alone.

What is appointment data worth? No public price exists for appointment or booking records alone. Company records sell at a median of $72.5K per company a company, with published offers at a median of $278K per company, and the bookings count for more when they are joined to the work they led to.

Can a clinic sell its appointment records? Only de-identified, or with each patient's authorization. For a provider HIPAA covers, appointments are protected health information, which it cannot sell without authorization; data that meets HIPAA's de-identification standard is no longer protected.

Is my booking software already training AI on my data? Some say so in their terms. Housecall Pro's terms record consent to training its AI models, Square's licence includes training models, and Mindbody trains on de-identified or aggregated data. ServiceTitan commits not to train models for third parties on its customers' data.

What this means for you

Booking and scheduling records sell when they show work being done, and a booking system rarely stops a business licensing them. What decides a sale is what the records show and who is in them:

1. Name what you hold. Jobs and dispatch with notes and invoices, or appointments alone. The first has named buyers; the second sells only as part of a fuller record.
2. Read your system's terms for ownership, the vendor's licence, aggregated data, its AI training and, for a franchise, the franchisor's rights.
3. Export while the account is live, through the system's own export or API, never a scraper.
4. Take the people out, customers, patients and staff alike, including addresses and photos, and leave out health and children's records you cannot stand behind.
5. Value your data with bookings or scheduling among your systems, and check the buyer board for the buyers that name your kind of business. The pages for contractors and trades and home improvement companies show what buyers ask of each.

AIData.Broker acts as the holder's agent: the holder itself licenses its data to the buyer, and only the holder pays us, a fee fixed in the mandate and due only if a sale closes. We run sales for holders based in the United States, the United Kingdom and Canada. For the wider picture, see how to sell company data to AI companies and what a data licensing agreement contains.

This guide is research from public sources, not legal advice.