AI in the Trades: Booking, Quotes and Service Agreements (2026)
This guide is about the written work around the job in a trades business: enquiries, booking, quotes, service agreements and follow-up. Here are workflows to try out, where a human approves, and what you need to have in order.
Table of contents
In short: For a trades business, this guide is about how AI (artificial intelligence) can help with the written work around the job: sorting enquiries, suggesting times, drafting quotes, keeping track of service agreements and following up. The master craftsman or the office approves before anything goes to the customer. Start with one workflow you can measure, e.g. follow-up on sent quotes.
Why tradespeople are looking at AI
If quotes, customer replies and rescheduling of appointments take up time after the working day, you can look into one of the workflows below.
That's where AI can help: not with the craft itself, but with the administration around it.
What can AI concretely do for a trades business?
Sort and answer written enquiries
Check whether your chosen solution can read incoming emails and forms, sort them by type (new job, emergency damage, question about an existing agreement) and draft a reply based on the business's own information: opening hours, area, what you do and what you don't do, with the access, licence and setup you have. Emergency cases should be flagged so they come first.
Booking and rescheduling appointments
Check whether the solution can suggest available times from your calendar and help reschedule an appointment. The office or the master craftsman must approve the booking before the time is registered and the confirmation sent.
Draft quotes
Try out in a pilot whether AI can, from the customer's description, photos and your own price lists, produce a usable first draft of a quote that the master craftsman reviews and corrects. The price and the assessment of the job are always the master craftsman's. Also check whether the solution can keep an eye on quotes without a reply and prepare a draft of a friendly follow-up.
Service agreements: an example
Service agreements (e.g. boiler inspections, heat pump servicing and annual check-ups) are regular work, and therefore suit a fixed workflow:
- The basis: each installation has a date of last service and the interval stated in the contract with the customer.
- The reminder: when the next visit approaches, the agent prepares a draft reminder with suggested times that fit the calendar.
- The customer's reply: the customer chooses a time or asks for another.
- The booking: the office or the master craftsman approves before the time is put in the calendar and confirmed to the customer.
The interval is what the contract says, and a human approves every reminder and every booking. The workflow makes it easier to keep track of agreements, but it doesn't relieve the business of keeping an eye on them.
History per customer and installation
When emails, service reports and photos are gathered per customer, and preferably per installation, check whether AI can help find what otherwise lies scattered in the inbox: when was a part last replaced, and what did the fitter write last time? It requires customer data to be properly recorded, e.g. that the same address doesn't appear three times in three different spellings.
What AI should not do in a trades business
- Set prices alone. AI can suggest a starting point from your price lists, but the final price is the master craftsman's.
- Promise times or solutions that nobody has approved. Messages to customers are drafts until a human has seen them.
- Assess technical questions. Whether an installation is in order, or what a fault requires, is the tradesperson's assessment.
- Cold-call customers without knowing the rules. It is unsettled whether a conversational AI voice is an automated calling system under section 10(1) of the Danish Marketing Practices Act. The other rules on telephone sales and personal data continue to apply. Read the walkthrough in Are AI calls legal in Denmark?.
The phone
If enquiries come in by phone, first consider the simple option: that missed calls become a task with the number and time, so the office can call back. Check whether your phone system can do that. If you're considering a solution in which an AI answers calls, check how it states that it is an AI, how the conversation is processed, and whether the call can be transferred to a human.
How to get started
- Choose one workflow that takes up a lot of time, e.g. follow-up on quotes or reminders about service agreements.
- Measure the starting point: how many quotes are without a reply after a week, and how many service agreements are not renewed on time?
- Run a pilot for a few weeks in which everything the AI does is approved by a human.
- Measure again, and decide from your own figures whether the workflow should be expanded, adjusted or dropped.
Typical pitfalls
- Starting with the hardest thing. Begin with a workflow that is regular and easy to measure, not with emergency damage that requires a technical assessment straight away.
- Data that doesn't fit together. If the same customer appears several times in the calendar, order system and email, check whether duplicates give wrong or incomplete history. Clean up first.
- Messages without an owner. Decide who in the office approves the drafts, and when in the day it happens, so they don't just sit there.
- Too many changes at once. Let the tradespeople get used to one new workflow before the next one arrives.
What it costs, and what to ask
The price depends on which systems the AI has to talk to (calendar, email, quoting and order system), how many enquiries you have, and whether the solution is charged by users or by usage. So ask:
| Question | Why it matters |
|---|---|
| Can it talk to our calendar and order system? | Clarify which information still has to be transferred manually |
| Do we approve before anything goes to the customer? | It's your name on the message |
| Where is the customers' data processed, and by whom? | If the supplier processes personal data on your behalf, a data processing agreement is required |
| What does it cost at our volume of enquiries? | The price can depend on users, usage and integrations |
| Can we stop again and take our data with us? | You shouldn't be locked in |
Customer data and rules
Customers' names, addresses and descriptions of jobs are personal data when the customer is a private individual. If they are processed by an AI solution, there must be a lawful basis, customers must be informed about the processing, and if the supplier processes the information on your behalf, you must have a data processing agreement. Limit the information to what is necessary for the job.
If you'd like help finding the first workflow where AI can take over some of the written work, get in touch.
Sources
- The Danish Consumer Ombudsman, Overview of the rules on telephone sales: https://forbrugerombudsmanden.dk/media/56664/oversigt-over-reglerne-for-telefonsalg.pdf (accessed 12 October 2026)
- The Danish Marketing Practices Act (Lov om markedsføring), consolidated act no. 1420 of 2 December 2024: https://www.retsinformation.dk/eli/lta/2024/1420 (accessed 12 October 2026)
- The Danish Data Protection Agency, "You are also responsible when you share" (data processors): https://www.datatilsynet.dk/regler-og-vejledning/gdpr-univers-for-smaa-virksomheder/trin-7-du-er-ogsaa-ansvarlig-naar-du-deler (accessed 12 October 2026)
- The Danish Data Protection Agency, the basic principles: https://www.datatilsynet.dk/regler-og-vejledning/grundlaeggende-begreber/hvad-er-dine-forpligtelser/de-grundlaeggende-principper (accessed 12 October 2026)
Frequently asked questions
How can tradespeople use AI?
This guide is about the written work around the job: sorting enquiries, booking, draft quotes, reminders about service agreements and follow-up. A human approves before anything goes to the customer.
Can AI make quotes for a tradesperson?
Try out in a pilot whether AI can produce a usable first draft from the customer's description and your own price lists. The master craftsman reviews, corrects and sets the price.
Can AI book customers into the calendar?
Check whether the solution can suggest available times. The office or the master craftsman must approve the booking before the time is registered and the confirmation sent.
Can AI answer the phone for a tradesperson?
This guide is about the written workflows. If you're considering a solution in which an AI answers calls, check how it states that it is an AI, how the conversation is processed, and whether the call can be transferred to a human.
What is the first thing a trades business should automate?
Start with one workflow that is easy to measure, e.g. follow-up on sent quotes or reminders about service agreements.
#trades #booking #ai customer service #automation #agent360
Contents
How many of your enquiries are the same ten questions?
We look at what can genuinely be answered automatically for you, and what cannot. An honest answer, even if it is not much.
Book a reviewHow many of your enquiries are the same ten questions?
We look at what can genuinely be answered automatically for you, and what cannot. An honest answer, even if it is not much.
How we built it · Email & prioritisation
The inbox, sorted
An inbox where AI sorts the emails, writes the reason next to each email and can draft a reply based on the customer’s case.
- Email in
- AI sorts
- You reply
Related articles

AI in the Call Centre: How to Handle Peak Loads Without Hiring More People
AI in the call centre: handle peak loads, follow-up work and knowledge search without losing quality. Frameworks, rules and measures before you build.

AI for Car Sales & Leasing: Filling the Test-Drive Calendar
AI for car dealers and leasing companies: faster replies, follow-up on expiring leases and old leads, test-drive booking and service. Done with control.

AI for Events and Courses: Registration, Reminders and Delivery
How AI can support administration around events and courses: registrations, waiting lists, reminders, the instructor pack and the rules for invitations.