Key takeaways
- Use AI writing tools for words, not numbers: turning survey notes into clear scope, exclusions and customer-friendly explanations.
- Construction lags on adoption: 6.7% of UK construction businesses used AI text generation in autumn 2026, against 19% of all businesses[1].
- Most firms using AI check its output, and two-thirds check it significantly[2]; for quotes, every figure and every compliance statement must be checked by you.
- Give the tool your template, your standard inclusions and your survey notes. Never let it invent products, prices or regulations.
- Keep personal data out of general-purpose tools unless you know how that data is handled.
Where AI helps with quotes, and where it does not
Writing quotes is one of the most disliked jobs in a plumbing and heating business. After a day on the tools, turning scribbled survey notes into a clear, professional document is slow, and quotes that go out late or vague lose work. AI writing tools are good at exactly this kind of task: taking rough notes and producing tidy, consistent text.
They are not good at knowing your prices, your costs, the property or the regulations. They produce confident text whether or not it is right. The safe division of labour is simple: the tool drafts the words, you supply and check every number, product and technical claim.
| Task | Good use of AI? | Why |
|---|---|---|
| Turning survey notes into a scope of works | Yes | Structure and clarity from your facts |
| Writing exclusions and assumptions clearly | Yes | Consistent wording across quotes |
| Explaining a technical item in plain English | Yes, then check | Useful drafts, but technical accuracy must be verified |
| Setting prices or labour hours | No | Needs your costs and job history |
| Choosing products or boiler outputs | No | Needs survey data and manufacturer information |
| Stating regulations or grant amounts | No, unless you verify each one | Tools can be out of date or wrong |
Construction has been slow to adopt
The trade is cautious, for good reason. In autumn 2026, 81% of UK construction businesses said they did not use any AI technology[3]. Only 6.7% used large language model text generation, against 19% of all UK businesses[1].
Source: Office for National Statistics, 2026[1]
Show the data
| Item | Value |
|---|---|
| Construction businesses | 6.7% |
| All businesses | 19% |
Among UK businesses that do use AI, text generation is by far the most common type, used by 85% of them[4], and administration is one of the main areas of use[5]. Quote writing is administration, so it is a natural place to start, as long as it is done carefully.
Feed the tool the right inputs
Your quote template
Paste your standard structure: summary, scope, inclusions, exclusions, assumptions, programme, payment terms, warranty. Ask the tool to follow it exactly.
Your standard wording
Give it your usual inclusions and exclusions for that job type, so it adapts your words rather than inventing new ones.
The survey facts
Your notes: boiler position, flue route, gas supply findings, condensate route, system condition, controls, access. Facts only, no customer details it does not need.
Clear instructions
‘Write a scope of works from these notes using our template. Do not add any products, prices, regulations or figures that are not in the notes. List anything unclear as a question for me.’
A review pass
Ask it to list every assumption it made. Then read the whole draft yourself.
The instruction to flag unclear items matters. A tool told to ask questions will often spot gaps in your notes, such as a missing condensate route, before the customer does.
Never let it invent
Text generators produce plausible sentences. Plausible is not the same as correct, and in a quote an invented detail can become a promise. What you say or write to a customer about the service can become part of the contract if they rely on it, and where no price is agreed, a customer only has to pay a reasonable one[6].
- Every price, quantity, output and duration comes from you, not from the tool.
- Every product name and model number is checked against your merchant quote.
- Every statement about regulations, registration, notification or grants is checked against the official source or removed.
- Warranty terms come from the manufacturer's documents, not from memory or the tool.
- Nothing promises a result you cannot guarantee, such as a specific energy saving.
Businesses already using AI tend to know this: 84% reported at least some human checking of AI outputs, and 67% reported significant checking[2]. For customer-facing quotes, significant checking is the minimum.
A post-drafting checklist
- Does the scope match what you saw on the survey, item by item?
- Are all figures yours, and do the totals add up?
- Are inclusions and exclusions consistent with each other?
- Is anything promised that you do not normally do?
- Are regulations and grants stated correctly, or not at all?
- Would a customer with no technical knowledge understand every sentence?
- Does it sound like your firm, or like a brochure?
If the answer to the last question is ‘brochure’, ask the tool to rewrite in plain, short sentences, or edit it yourself. Customers trust quotes that sound like the person who came to their house.
Set house rules before the team uses it
If more than one person writes quotes, agree how AI tools are used before habits form. A short written policy avoids the situation where one engineer's quotes are checked carefully and another's go out unread.
- Which tools may be used for business documents, and with which account settings.
- What may be pasted in: survey facts and standard wording, yes; customer names, contact details and payment information, no.
- Who checks AI-drafted quotes before they are sent, and against what checklist.
- Where approved standard wording is kept, so everyone starts from the same inclusions and exclusions.
- What to do if an error is found after a quote has gone out: correct it in writing straight away.
Lack of expertise is the barrier construction businesses cite most often when they have not adopted AI[7]. A one-page policy and an hour of practice on old quotes is usually enough expertise for this task.
Other quote paperwork it can draft
Once you trust the process for quotes, the same approach works for the rest of the paperwork around a job: variation descriptions written up from a photo and a sentence, quote follow-up messages in your own tone, plain-English explanations of why a system clean or gas pipe upgrade is needed, and handover notes listing what was fitted and how to use the controls. Each one follows the same rule: you give the facts and figures, the tool drafts the words, and you read every line before it reaches a customer.
Keep customer data safe
Quote drafting needs survey facts, not personal details. Leave out names, phone numbers and full addresses when you paste notes into a general-purpose tool, and add them back in your own quote template. Check the tool's settings and terms for how your inputs are stored and used before you rely on it for business documents.
Treat links and attachments that claim to come from AI or software providers with the same caution as any other email. Phishing was the most common type of breach among UK businesses that had a breach or attack in 2025, experienced by 85% of them[8].
Example: before and after
Example (fictional) survey note: ‘combi kitchen cupb, flue horiz rear, ok clearances. 15mm gas from meter approx 9m, 4 elbows, prob undersize for 30kW, upgrade to 22. condensate int to sink waste. sludge in rads, mag filter none. wants smart stat.’
Example draft after AI, checked and edited by the engineer: ‘Remove the existing boiler from the kitchen cupboard and fit the new combination boiler in the same position, with a new horizontal flue through the rear wall. Replace the gas supply pipe from the meter to the boiler with larger pipework so the new boiler receives the gas supply it needs. Connect the condensate pipe internally to the kitchen sink waste. Clean the heating system with a chemical cleaner and fit a magnetic filter. Supply and fit a smart thermostat and set it up with you on the day.’ Prices, model numbers and the warranty term were then added from the firm's own build-up.
Measure whether it helps
- Time from survey to quote sent, before and after.
- Number of customer questions about what a quote includes.
- Errors found at the checking stage, by type.
- Quote win rate, which often improves simply because quotes go out faster.
Use the tool for the blank page, not for the decisions. You are still the one signing the quote.
Frequently asked questions
Can AI write my plumbing quotes for me?
It can draft the words: scope, exclusions and plain-English explanations from your survey notes. Prices, products and compliance statements must come from you and be checked.
Is it safe to put customer details into an AI tool?
Leave out personal details where you can and add them in your own template. Check how the tool stores and uses what you enter.
How many trades businesses use AI?
Relatively few. In autumn 2026, 81% of UK construction businesses used no AI at all[3], and 6.7% used text generation[1].
What should I check in an AI-drafted quote?
That the scope matches the survey, every figure is yours, inclusions and exclusions agree, regulations are correct or absent, and the wording is plain.
References
Office for National Statistics (2026). Business insights and impact on the UK economy: wave 165 dataset (AI tables) (opens in a new tab) ↑
Department for Science, Innovation and Technology (GOV.UK) (2025). AI adoption survey report 2025 (opens in a new tab) ↑
Office for National Statistics (2026). Business insights and impact on the UK economy: wave 165 dataset (AI tables) (opens in a new tab) ↑
Department for Science, Innovation and Technology (GOV.UK) (2025). AI adoption survey report 2025 (opens in a new tab) ↑
Department for Science, Innovation and Technology (GOV.UK) (2025). AI adoption survey report 2025 (opens in a new tab) ↑
legislation.gov.uk (The National Archives) (2015). Consumer Rights Act 2015, section 51 (opens in a new tab) ↑
Office for National Statistics (2026). Business insights and impact on the UK economy: wave 165 dataset (AI Prevent or Delayed) (opens in a new tab) ↑
Department for Science, Innovation and Technology (GOV.UK) (2025). Cyber security breaches survey 2025 (opens in a new tab) ↑