AI at Work • For plumbing, HVAC and electrical companies
The Call You Missed at 9:40 PM: After-Hours Response for Plumbing, HVAC and Electrical Companies
Plumbing, HVAC and electrical companies lose their most urgent, best-paying calls after hours. How to answer them, where AI fits safely, and what to track.
By Justin Beegel · · 7 min read
If you run a plumbing, HVAC or electrical company, you already know how this goes. The phone sits there quiet at 2 in the afternoon. Then it rings at 9:40 at night because a water heater gave out, or the AC quit in August, or half the house just went dark.
Those are the most urgent calls a homeowner will ever make to you. A lot of the time they’re also the best jobs you’ll book that week. And they’re the ones most likely to hit voicemail.
Somebody with water on the floor doesn’t leave a message and sit tight. They hang up and call the next company on Google. Whoever picks up first and sounds like they know what they’re doing usually gets the job. Often they get the customer for the next ten years too.
Why the usual setups come up short
Most shops end up with one of three arrangements. I’ve seen versions of all of them, and each one breaks in a predictable spot.
- The owner’s cell phone. Nobody has better judgment than you do, but it doesn’t scale and it eats your evenings. It also fails the first night you’re asleep, driving, or up to your elbows in somebody else’s job.
- A rotating on-call tech. Better coverage. The catch is that technicians aren’t dispatchers. Calls get answered from a crawlspace, details get scribbled on a glove, and price conversations happen before anybody has the facts.
- A traditional answering service. Someone always picks up, which counts for something. But that operator is reading a script for dozens of companies. They can take a name and a number. They usually can’t tell a real emergency from a Tuesday-morning job, and they can’t see your schedule or book anything on it.
The problem is the same in every case. Nobody on the line has your company’s know-how and the customer’s details in front of them at the same time.
What good after-hours response looks like now
The goal is easy to say out loud. Every after-hours call gets answered right away. Somebody (or something) figures out what’s actually going on and sorts it. The right next step happens, and nothing gets lost before morning.
A few years ago that was only realistic if you could afford a night dispatcher. AI makes it doable for a ten-truck company. Here’s roughly how it’s put together.
Answer every call as your company
A voice or text assistant picks up in a ring or two. It says up front that it’s your after-hours line and that it’s automated. Then it asks what a sharp dispatcher would ask. What’s happening? Is water running right now? Is anybody in the house elderly, a baby, or medically fragile and without heat or AC? Do you smell something burning, or see an outlet sparking? What’s the address?
What makes this different from a script is that it actually follows the answers. “The water heater is making a popping noise” and “there’s water coming through the ceiling” are both plumbing calls. Only one of them means you’re rolling a truck tonight.
Sort calls by your rules
You already have rules. They might only live in your head, but they’re there. You know what counts as an emergency and which zip codes you’ll drive to at night. You know the after-hours fee. You know when to tell someone to shut the main off, or to hang up and call the utility.
Getting those rules written down is most of the work here. Honestly, it pays off well beyond the phone. Once they’re written, the assistant applies them the same way at 9:40 PM as it does at 3 AM, which is more than most people can say about themselves at 3 AM.
Do the right thing next
What happens next depends on the call. Usually it lands in one of these buckets.
- Safety comes first. That means clear steps for shutting off the water, the gas or a breaker, or a flat instruction to call 911 or the utility when it’s bigger than a service call.
- A true emergency gets the on-call tech paged with a real summary: address, problem, gate code or access notes, and what the customer said in their own words. That beats a missed call and a garbled voicemail.
- Urgent stuff that can wait until morning gets booked into the first open slot tomorrow and confirmed by text. The customer goes to bed knowing somebody’s coming, which is rare at that hour.
- Routine requests get logged, the common questions get answered, and the whole thing goes into the morning queue with enough detail that your office only has to call back once.
Hand it off clean in the morning
By 7 AM your office should see every overnight contact in one place. What came in, how it got sorted, what was promised, and what still needs a human. That morning summary is sneaky valuable. It’s also a nightly record of demand you’ve probably been guessing at for years.
The same call, from the homeowner’s side
Picture it (this is a made-up example, but you’ve lived some version of it). It’s 9:40 PM and water is spreading across a laundry room floor. The homeowner calls the first company they find.
The after-hours assistant picks up on the second ring. It asks what’s going on, hears “active leak,” and walks them through finding and closing the main shutoff while they’re still on the phone. Once the water’s off, it confirms the address and notes the leak is coming from the water heater. It checks that the address is inside the after-hours service area, states the after-hours fee plainly, and pages the on-call plumber with a clean summary.
A few minutes later the homeowner gets a text with the plumber’s name, an arrival window the plumber has confirmed, and a reminder not to turn the water back on.
Now compare that to a voicemail greeting and a callback that may or may not happen. People remember which company got them through a bad night. They tell their neighbors, too.
Writing down your emergency rules
Most of the value in this whole project comes from putting on paper the calls you already make by gut. A first draft usually answers these questions.
- What’s always an emergency? Active leaks, sewage backups, sparking or a burning smell. Also no heat when it’s below a temperature you pick, and no cooling above one (in South Florida, that second one does a lot of work).
- What’s urgent but fine until morning? A dripping fixture, a water heater that’s noisy but still working, one dead outlet.
- How far will you go at night? The zip codes or the drive time you’ll actually cover after hours.
- What does after-hours service cost? The fee, said the same way every single time.
- When do you send them somewhere else? A gas smell goes to the gas utility and 911. Downed lines go to the power company. Anything life-threatening goes to 911.
- Who’s on call, and how do you reach them? The rotation, the backup, and how long to wait before going to the next person.
Once you’ve got this written, you’ll use it everywhere. It helps daytime dispatch. It helps you train new techs. It makes the answers on your website better too.
Where people stay in charge
I wouldn’t let any automated system quote a final price on complicated work. I wouldn’t let it diagnose a gas smell, either, or promise an arrival time your on-call tech hasn’t signed off on.
The assistant’s job is to understand the problem, sort it, calm the person down, get the details and send it to the right place. Your technician’s job stays your technician’s job. Drawing that line clearly is what makes customers trust the thing, and it’s what keeps your techs from hating it.
Two more guardrails I’d insist on. A customer with a real emergency should always be able to get to a person. And callers should be told, right at the start, that they’re talking to an automated assistant. Being straight about that costs you nothing, and it protects your name.
What to track
A project like this should pay for itself in numbers you already understand. Before you change anything, write down where you stand for a few weeks. Then compare.
- How many after-hours calls got answered, versus going to voicemail or hanging up.
- How long it took to get a first response after hours.
- How many emergency calls turned into booked jobs, and what those tickets averaged.
- How many morning callbacks got done before 10 AM.
- How often the on-call tech got woken up for something that wasn’t an emergency.
What actually makes it work
The technology is the easy part at this point. I spent almost 18 years running an agency before NUDGE, and the tools were rarely what made or broke a project. The thinking behind them was.
Same thing here. A system that helps and one that drives customers nuts can run on the exact same software. The difference is whether somebody sat down and wrote out the emergency rules, the service area, the pricing language and the handoff to real people. That’s what a great dispatcher carries around in their head. You write it down once and it works for you at every hour of the day.
If you want a rough idea of what faster response could be worth to your shop, the AI Opportunity Calculator walks you through it with your own numbers.
Questions owners ask
Will customers be annoyed by an automated after-hours assistant?
They’re a lot more annoyed by voicemail. What matters is that the assistant says up front it’s automated, understands what they tell it, gives them clear next steps, and gets a real person involved fast when it’s a true emergency.
Can an AI assistant book jobs directly into my schedule?
Yes, if your scheduling or field service software allows it. Most companies start by booking non-emergency calls into the first morning slots and paging the on-call tech for emergencies. You set the rules for both.
What should an after-hours assistant never do?
It shouldn’t try to diagnose a gas leak or an electrical hazard beyond giving basic safety steps and sending people to 911 or the utility. It shouldn’t quote final prices on complicated work. And it shouldn’t promise an arrival time your on-call tech hasn’t confirmed.
Keep reading
AI at Work · · 7 min read
The Front Office Is the Bottleneck: Where AI Fits in a Dental Practice or Law Firm
A dental office and a law firm don’t look alike, but they get stuck in the same place: the front desk. Where AI can take real work off your staff, and where confidentiality and professional judgment should keep it out.
AI at Work · · 7 min read
The First Good Quote Wins: AI for Distributors and Supply Houses
Quote requests show up as emails, PDFs, photos and forty-line spreadsheets full of part numbers and substitutes. Where AI can shorten the trip from request to quote, while your people keep control of pricing.
AI Visibility · · 7 min read
How AI Assistants Decide Which Local Business to Recommend
An AI assistant builds its local recommendations from evidence it can find and cross-check. A plain-English look at that evidence, and the order I’d fix it in.
Want this thinking applied to your business?
Tell me what is slowing the business down, and I will tell you where I would start.
