Home Services AIChat and voice assistants

An assistant that knows your business.

We build chat and voice assistants that answer customer questions from your own content, take bookings and capture leads. When a conversation needs a person, it goes straight to your team.

Start a project See how it works

Replies after hours
Bookings taken automatically
Clean handover to staff

Call summary

Example

Caller asked about availability for a boiler service next week. Booked Thursday morning and sent a text confirmation. They also asked about a warranty claim, so this has been passed to the office team with the details.

Checked by a person before it goes out

Customers ask the same questions all day, and calls or messages that come in after hours often go unanswered. We build assistants that answer from your approved content and pass anything sensitive or unclear to a person.

Where your assistant works.

One assistant can answer on several channels, using the same approved knowledge and the same rules for handing over.

Website chat

A chat window on your site that answers questions, suggests the right product or service and collects contact details for your team.

WhatsApp Business

Customers message you on WhatsApp and get answers, booking links and order updates, and your team can step in at any point.

Phone answering

A voice assistant answers calls, handles common questions and takes messages, using Twilio for calls and texts. Each call is logged with a written summary.

Bookings and lead capture

The assistant checks availability in your calendar or booking system and books appointments, or adds new leads straight to your CRM.

Handover to your team

Complaints, complex questions or anything outside its brief go to your team by Slack or email, with the conversation so far attached.

Answers from your content

Replies draw on your website, FAQs, policies and price lists, and the assistant is set up to hand over when it is unsure.

Done properly.

The details that decide whether chat and voice assistants pays off.

Checked against real questions first.

Before launch we run the assistant through real customer questions, including awkward and off-topic ones. A person reviews every answer, and we adjust the content and rules until you are happy to sign it off.

Assistant test plan

Working

Collect real customer questionsDone
Load approved contentDone
Test awkward and off-topic questionsDone
Check handover to the teamDone
Review replies with youLive
Go live on websiteNext

Your customers' data, handled with care.

Conversations are stored in line with UK GDPR, and we use business API settings where your data is not used to train models. Customers are told they are talking to an AI assistant, and sensitive details can be kept out of it entirely.

After-hours call

CallCustomer rings after hours

TaskAssistant answers question

SplitBooking or needs person

SMSBooking confirmation sent

ReportSummary to your inbox

How it works.

Clear steps, and you approve each one before we move on.

01
Gather
We collect your FAQs, policies, prices and real customer questions.
02
Build
We set up the assistant, its channels and the rules for handover.
03
Test
A person reviews answers to real questions before anything goes live.
04
Improve
We read conversation logs each month and fill gaps in its knowledge.

Example project

A heating company missing calls while on jobs

A small heating company misses calls while engineers are out and the office is busy. Many callers want simple things such as prices, availability or a booking, and some never call back.

What we would do

Gather common questions, prices and booking rules
Set up a voice assistant on the main number
Connect it to the job booking calendar
Route urgent or complex calls to the office

An illustration of how we would approach it, not a client result.

What you would get

A phone assistant for missed calls
Bookings added to your calendar
Written call summaries by email
A monthly review of conversations

Start a project like this

Related work.

Projects where we did this kind of work.

Own product · iOS and Android app

RepCam AI

A fitness app that counts your reps through the camera, with challenges and daily drops.

Read the case study

Own product · Web app

InfinityFX

Forex market analysis with AI-written explanations and Stripe subscriptions.

Read the case study

More workSee everything we've built.Client projects and our own products, with what we did on each.See all work

Questions, answered.

Will customers know they are talking to AI?

Yes. We recommend the assistant says so at the start, which is good practice and builds trust. Customers can ask for a person at any time and the assistant passes them on.

What happens if the assistant gets something wrong?

It is set up to answer from content you approve and to hand over when it is unsure. A person reviews conversation logs regularly, and we update its knowledge and rules whenever we spot a weak answer.

Which tools do you use to build assistants?

We use OpenAI and Claude models, Twilio for phone calls and texts, and WhatsApp Business for messaging. We connect them to your calendar, CRM or booking system so the assistant can act on what it hears.

How much does an AI assistant cost?

The build is quoted after a short call, with a fixed price agreed before we start. Hosting, model usage and monthly improvements then run on a fixed monthly fee.