Conversational AI for Trades and Home Services: Capture Every Job, Quote From Photos, Get Paid Faster

Plumbers, electricians, builders and cleaning companies win or lose work in the first few minutes of an enquiry, often while their hands are full on another job. WhatsApp and conversational AI can catch those enquiries, collect the details and photos, and keep the customer informed until the invoice is paid. This guide shows how, and where a person still needs to step in.

Which conversations fill a tradesperson's day?

Most trades messages fall into five groups: availability, prices, arrival times, invoices and follow-up work. Typical examples:

  • "Can you come today?"
  • "How much for a new socket in the kitchen?"
  • "What time will you get here?"
  • "Can you send the invoice again?"
  • "While you're here, could you look at the outside tap?"

Most of these are predictable and need the same information each time: what the job is, where it is and how urgent it is. That makes them well suited to automation, leaving the tradesperson to focus on the work and on the jobs that need judgement.

Which channels do customers use?

Customers mostly use the phone, followed by WhatsApp, web chat on your site and Facebook. Each has a role:

  • The phone is the most important and the most often missed, because you are up a ladder or under a sink.
  • WhatsApp suits sending photos, addresses, quotes and updates.
  • Web chat catches visitors reading your services page.
  • Facebook matters for local recommendations and community groups.

When all of these arrive in one shared inbox, the office or the owner can see every job enquiry in one place, instead of scrolling through several apps at the end of the day.

Which automations make the biggest difference?

The automations that make the biggest difference are missed-call follow-up, job intake with photos, quotes with an accept button, scheduling updates and invoicing by link.

  • Missed-call follow-up. A missed call triggers a WhatsApp message inviting the caller to describe the job.
  • Job intake. The customer sends photos or video, the address and a short description.
  • Quotes with an accept button. The customer approves in one tap, giving you a written record.
  • Scheduling and "on my way" messages. The customer knows when to expect you.
  • Invoice and payment link. Sent as soon as the job is done.
  • Review requests. A short message after the job, sent once.
  • Seasonal reminders, with consent. Boiler servicing or gutter cleaning, only for customers who agreed.

Cleaning companies have their own set: recurring bookings, access instructions for each property and a message telling the customer which cleaner has been assigned. Start with missed-call follow-up and job intake, as they bring in work most directly.

What should an AI agent handle, and what should people handle?

An AI agent should collect job details, classify urgency and book visits into available slots, while people handle quotes on complex jobs and emergencies. The agent asks the same questions every time: what the problem is, where the property is, when the customer is available and whether anything is urgent. It can then offer a slot from your calendar.

A rewiring job, an extension or anything unusual needs a person to quote. Emergencies, such as a gas smell or a serious leak, go straight to a person, and the agent should never give safety advice of its own. The agent should say that it is automated, and customers should always be able to ask for a person.

How can voice help when you are on site?

An AI voice agent can answer the phone while you are working, so calls do not go unanswered. It greets the caller, takes the job details, flags urgent jobs and follows up on WhatsApp for photos. For the full picture, see the AI phone answering guide.

What compliance points matter for trades?

The main compliance points for trades are protecting access details and getting marketing consent.

  • Door codes and keys. Do not leave alarm codes, key-safe numbers or key locations lying around in plain chat notes. Store them securely, limit who can see them and remove them when the job or contract ends.
  • Marketing consent. Job updates and invoices are service messages; seasonal reminders and offers need the customer's agreement and an easy way to stop them. See the opt-in and opt-out guide.

Customer names, addresses and access details are personal data under the General Data Protection Regulation (GDPR). This is general information, not legal advice.

What does a full job look like?

A full job runs from a missed call to photos, an urgent flag, a confirmed visit, an "on my way" message, payment by link and a review. For example:

  1. At 7am, a customer finds water leaking under the kitchen sink and calls a plumber.
  2. The plumber is on another job, so the call is missed.
  3. An automatic WhatsApp message asks the customer to describe the problem and send photos.
  4. The customer sends two photos and the address. The job is flagged as urgent.
  5. The plumber sees the flag, checks the photos and confirms a visit that morning.
  6. An "on my way" message tells the customer the plumber is 20 minutes away.
  7. After the repair, the invoice and a payment link arrive in the chat, and the customer pays.
  8. The next day, a short message asks for a review.

A call that would once have been lost became a paid job, with the plumber only stepping in to confirm the visit and do the work.

How do you measure success?

Measure success with three numbers: missed calls recovered, quote acceptance and time to payment.

  • Missed calls recovered is the share of missed calls that turn into a conversation or a job.
  • Quote acceptance is the share of quotes the customer approves.
  • Time to payment is how long it takes from finishing a job to being paid.

Record each one before you change anything, then compare month by month to see which automations bring in work and cash, and which need adjusting.

How Arino One fits

Arino One runs WhatsApp, SMS, RCS, email, voice, live chat and 18+ other channels in one dedicated instance the client owns, built on Arino Core. Infobip sits behind every instance on the client's own account, billed at cost with no markup, and any AI runs on the client's own Claude or OpenAI key. Customers can start with a free remix via client.arino.one, choose done-for-you setup from $350 one-time, or run on Arino Core from $50 a month with the first month free. To talk it through, email hello@arino.one or use the contact page.

FAQ

Can WhatsApp help a tradesperson win more jobs?

Yes. Many jobs are lost simply because nobody answered in time. WhatsApp lets you reply to missed calls automatically, collect job details and photos, and send quotes customers can accept with one tap, so the customer does not move on to the next name on the list.

Can customers send photos for a quote?

Yes. Customers can send photos or a short video of the problem, along with their address, in the same chat. For simple jobs that is often enough for a quote; for complex ones it helps you decide whether a site visit is needed and what to bring.

What happens to calls I miss while on a job?

An AI voice agent can answer the call, take the details and flag anything urgent. If a call is still missed, an automatic WhatsApp message can follow up straight away, inviting the caller to describe the job and send photos.

Can I send invoices on WhatsApp?

Yes. You can send the invoice and a payment link in the same chat as soon as the job is finished. Customers can pay from their phone, which usually shortens the time you wait to be paid.

Can an AI handle emergency callouts?

An AI agent can recognise that a job sounds urgent, collect the key details and alert a person immediately. The decision to attend, and any advice about safety, should come from a person. If there is a risk to life, the customer should contact the emergency services.

Next step

Apply this to your own deployment

This guide describes decisions we make on live instances. Tell us your channels, systems and region and we will map it to an architecture outline, a provisioning plan and an indicative commercial model — usually within one business day.