How We Built a 24/7 AI Receptionist for a Plumbing Business (And What We Learned)
By Chinemelu Ezeh, Founder & CEO at Amatou Technologies
When a home services company approached us, the problem was clear. Their salesperson was drowning. Potential customers were flooding them with calls and emails about pricing, availability, and service areas. The person who should have been closing deals was spending most of their day just answering routine enquiries. What followed taught us more about AI deployment in small businesses than any enterprise engagement ever has.
Here's what we built, what surprised us, and the honest lessons we took away.
The Problem
Plumbing businesses live and die by converting enquiries into booked jobs. Our client had no shortage of leads. Their Google presence and reputation were driving a steady stream of inbound calls and emails from potential customers. But their salesperson was spending the bulk of their day answering the same questions over and over. "Do you cover my area?" "How much does a boiler service cost?" "Can you come this week?"
Every hour spent fielding routine calls and replying to repetitive emails was an hour not spent following up with qualified leads, closing jobs, and building customer relationships. The salesperson was essentially working as a receptionist. The business was leaving revenue on the table because of it.
What We Built
We deployed a RAG-powered AI agent that handles both voice calls and email inbound queries. The agent is trained on the company's service catalogue, pricing guidelines, and service areas. Critically, it's also trained on the conversational scripts of their best-performing team members.
The agent can:
- Answer inbound calls with natural voice conversation
- Understand and triage the customer's issue (emergency vs. routine)
- Book appointments directly into their job management platform
- Send confirmation emails and follow-ups automatically
- Handle email enquiries with the same knowledge base
Lesson 1: It's Not for Every Plumber
This is the most important thing we learned, and we say it openly. An AI receptionist is not the right solution for every plumbing business.
If you're a solo plumber doing 3-4 jobs a week, you probably don't need this. A good voicemail and a quick callback routine will serve you fine. AI agents deliver clear ROI when there's a genuine volume problem. If your team is spending more time answering routine questions than actually selling, that's when it makes sense.
We learned that being upfront about this actually builds more trust than overselling. It's a lesson we now bake into every discovery call. Before we propose a solution, we establish whether the business has a clear ROI case and a sales process that benefits from automation.
Lesson 2: Voice AI Has Come Further Than People Think
One of our initial concerns was whether customers of a plumbing company would actually accept speaking to an AI. Would they hang up? Would they get frustrated?
The data surprised us. Completion rates for AI-handled calls were significantly higher than we expected. Voice AI has improved dramatically. Today's models handle natural speech patterns, interruptions, and varied accents far better than even a year ago. Customers care about getting their problem solved quickly, not about whether the voice on the other end is human.
The apprehension that tradespeople's clients won't accept AI may be outdated. What matters is the quality of the interaction, not the nature of the responder.
Lesson 3: Integration with Job Platforms Like Jobber Is Very Possible
A critical requirement was that the AI agent couldn't just take messages. It needed to book real appointments in the company's existing workflow. They use Jobber, one of the most popular field service management platforms.
We integrated the agent directly with Jobber's scheduling system. When a customer calls and needs a visit, the agent checks technician availability, proposes a time slot, confirms with the customer, and creates the job. All in real time. No double-booking, no manual data entry.
This kind of deep integration is what separates a useful AI agent from a fancy chatbot. The agent isn't just answering questions. It's executing business processes.
Lesson 4: Learn from the Best Closers in the Business
Here's something that made a measurable difference. We didn't just train the AI on factual information. We studied the scripts and conversational patterns of the company's best-performing team members. The people who converted the most enquiries into booked jobs.
We analysed how they greet callers, how they ask qualifying questions, how they handle objections about pricing, and how they create urgency without being pushy. Then we encoded those patterns into the AI agent's behaviour.
The result? The AI doesn't just handle calls. It handles them the way your best employee would. This is the real power of AI in customer-facing roles. You can take the behaviours of your top 1% and scale them across 100% of interactions.
Lesson 5: Simulate Your Customers Before You Go Live
Before launching the email inbound agent, we built a simulation framework that generated realistic customer personas and scenarios. Everything from a panicked homeowner with a flooded kitchen to a landlord requesting routine maintenance quotes for multiple properties.
Running hundreds of simulated conversations allowed us to rapidly identify edge cases, refine the agent's responses, and verify it handled the full spectrum of real-world queries before a single real customer interacted with it.
We now build a simulation layer to stress-test the AI before every deployment. It dramatically reduces the risk of embarrassing failures in production and gives clients confidence that the agent is ready.
Lesson 6: Establish Clear ROI and Sales Process Upfront
The biggest mistake we see businesses make with AI is treating it as a technology purchase rather than a business investment. Before we write a single line of code, we work with clients to define exactly what success looks like in numbers.
For this plumbing business, the maths was straightforward. Their salesperson was spending roughly 60% of their day on routine enquiries instead of closing. If the AI agent frees up even half of that time and the salesperson converts just a handful more jobs per month, at an average job value of £150-£300, the system pays for itself many times over.
We also learned to map out the full sales process with the client before deployment. Where does the AI agent sit in the customer journey? What happens after it books an appointment? Who follows up? Having these answers before launch means the AI integrates seamlessly into existing operations rather than creating a new set of problems.
The Results
The AI receptionist now handles inbound calls and emails 24/7. The salesperson got their day back. Instead of being buried in routine enquiries, they're now focused on what they were hired to do: closing jobs and building customer relationships.
More importantly, every enquiry is now handled professionally and consistently. It uses the exact approach that the company's best team members use, whether it comes in at 2pm or 2am.
Is This Right for Your Business?
If you're running a home services business and your sales team is spending more time answering routine calls and emails than actually selling, an AI agent might be exactly what you need. But it might not be. We'll tell you honestly either way.
Our process starts with a free discovery call where we assess whether AI automation makes sense for your specific situation. If the ROI is there, we'll show you a proof-of-concept before you commit to anything.