How to sell AI systems to plumbing companies

A plumbing company's worst hour is your best sales asset. At 9:40 on a Tuesday night a homeowner with water coming through a kitchen ceiling calls the first three plumbers on the map. Two go to voicemail. One answers and gets a job worth somewhere between a service call and a re-pipe. If you want to sell an AI system to a plumbing company, that is the whole pitch, and you can make it before you ever say the word AI.

This is the playbook for that sale. Why plumbing closes easier than most trades right now, what the owner is actually buying, what your demo has to do, what to charge, and when to walk away.

Why is plumbing an easier AI sale than most trades?

Because the owner already knows they are losing money on the phone, and they already know they cannot hire their way out of it.

The labor math is public. The Bureau of Labor Statistics counted about 510,600 plumbers, pipefitters and steamfitters in 2025, with median pay of $63,800 and employment projected to grow 7 percent through 2035. The number that matters for your pitch is the openings figure: roughly 42,000 a year, and BLS attributes many of them to replacing people who retire or leave the trade rather than to growth. The pipeline is thinner than the demand.

The trade association says the same thing from the employer's side. PHCC's 2026 environmental scan, published in January, cites National Association of Home Builders data showing roughly half of employers in trades including plumbing report difficulty finding skilled applicants. The same scan projects remodeling work up about 2 percent in 2026 and sewage and waste disposal construction up about 6 percent, with EPA announcing $3 billion in new funding plus $1.1 billion previously allocated for lead service line removal.

Read those two facts together and you get the shape of the sale. Work is coming. Bodies are not. Nobody in that position wants a sixth marketing vendor. They want the calls they already paid to generate to stop falling on the floor.

Plumbing also has a demand curve no receptionist covers. Dental practices get called during office hours. Plumbers get called when a water heater fails on a Sunday and at 2am during the first hard freeze of the year. The gap between the phone ringing and a human being available is wider here than in almost any appointment-based trade, and that gap is the product.

What is a plumbing owner actually buying?

Not AI. Four things, in this order.

Capture. Every inbound call answered, at any hour, with the caller's name, number, address and the words they used to describe the problem written down. This alone justifies the spend for most shops.

Triage. The difference between "no hot water" and "water is coming out of the wall" is the difference between a Thursday slot and a truck rolling tonight. A good agent asks the questions a dispatcher asks and tags the call accordingly. It also tells a panicking caller where the shutoff valve usually is, which is the most useful sentence anyone says in that conversation.

Booking. Scheduled work goes straight into the calendar with address and access details. Emergencies route to a human instead.

Recovery. The call that still gets missed, because the agent was on another line or the owner's cell was the number dialed, gets a text back inside a minute. Missed-call text-back is the least glamorous node in any of these systems and it is usually the one that pays for the whole thing.

Notice what is not on that list. No content generation, no review replies, no "AI strategy". Those exist and some plumbing shops use them. PHCC's scan cites a 2025 Housecall Pro survey of more than 400 home-service professionals in which over 70 percent had experimented with AI and roughly 40 percent of plumbing professionals use it regularly, mostly for content and administrative work. That is useful context for you, because it means your buyer has probably already typed something into ChatGPT and formed an opinion. It is not what you sell. Phone coverage is what you sell.

How do you open the conversation with a plumber?

With a question about last weekend, not a pitch about technology.

"How many calls came in after five on Saturday, and who answered them?" That is the whole opener. The owner either knows the number, which means they have already worried about it, or does not know it, which is worse and they will say so out loud. Either answer hands you the rest of the call.

What you should not do is lead with the technology. Pew Research surveyed 5,119 US adults in February 2026 and found 63 percent say AI is moving too quickly. The same survey found about half of adults now use an AI chatbot, up from a third in 2024, so this is not a room that has never heard of the category. It is a room that has heard plenty and is tired of the noise. A plumbing owner in their fifties who has been sold three CRMs and a lead-gen contract in the last two years has an immune response to the word AI, and you do not need the word to make the sale.

Two more things about this buyer. They take calls from the van, so your call is eight minutes long whether you planned for that or not. And they count jobs, not percentages. "This catches maybe four more jobs a month" lands. "A 19 percent lift in lead capture" does not.

If you have never run this call in a structured way, our demo call playbook has the twenty-minute sequence, including when to stop talking.

What should the demo actually do?

Hand them the phone and let them try to break it. Plumbing owners will try, which is good, because the ones who try are the ones who buy.

Before the call, load their business name and the plumbing pack, and run the live demo against their own domain so the website audit has something real to find. On the call, give them the link and go quiet. Five specific things to let them test:

1. The emergency script. Tell them to call in as a homeowner with a burst supply line. They are listening for whether the agent tries to book a Tuesday appointment instead of escalating. If it escalates and asks where the main shutoff is, they relax.
2. A price question in their own bands. Water heater swap, main line hydro jet, sewer camera inspection. They will know instantly whether the numbers coming back sound like their market or like a template.
3. The after-hours question. Ask it what happens if nobody answers. They want to hear a callback commitment with a time on it.
4. Missed-call recovery. Let them watch the unanswered call turn into a text and the text turn into a booking. This is the node owners point at.
5. Their own numbers in the ROI figure. Seed it with their average ticket and close rate. Arithmetic they supplied is arithmetic they cannot argue with.

Every published demo carries a visible sandbox disclosure saying the AI is a demo and that bookings inside it are simulated. Point at it rather than hoping nobody reads it. Plumbing owners are suspicious of vendors, and a vendor who volunteers the limits is a novelty. What the prospect sees on the page is covered in what your prospects actually see.

What does a good plumbing pack sound like?

Like a dispatcher, not a concierge.

The tell of a generic agent in this trade is warmth in the wrong place. A med spa agent should sound welcoming. A plumbing agent at 11pm should sound calm and fast, get the address early, and ask one question at a time. Our vertical packs carry the services, price bands and objections for each trade, and for plumbing that means the agent knows the difference between a clog and a backup, asks whether water is currently running, and does not try to upsell a maintenance plan to somebody standing in two inches of water.

Vocabulary that has to be right or the whole thing falls apart: shutoff, main line versus branch, hydro jetting, sewer camera, water heater versus tankless, slab leak, backflow test, permit-required work. Get one of these wrong in front of an owner and you are done for the call.

Language that has to be absent: firm quotes on anything that needs eyes on it. A good plumbing agent gives a band and books a diagnostic. An agent that commits to $285 for a job that turns out to be a slab leak has cost your client money and cost you the account.

What should you charge a plumbing company?

Retail is yours to set. Here is the cost side so you can do the math honestly.

The platform is $49 a month. That covers the agency-branded dashboard, the live demo site on your subdomain, the client portal, access to the fulfillment platform, and onboarding help, with 100 live voice minutes, 500 chat messages and 50 website audits a month for your own demos. There is no build fee on the custom software path, which is the part that changes how you sell.

Delivery is wholesale. Top Tier, which includes voice, is $500 a month and replaces about $745 a month of the same work bought line by line. Tier Two is $275 and Tier One is $225. Single services run $40 to $350 a month. Those are the prices you pay us, not the prices the client sees.

For a shop running three to six trucks, lead with Top Tier, because voice solves the problem you opened the conversation with. What you charge on top is a judgment call about your market and your nerve. Give the owner cost per recovered job rather than cost per month. If the system catches two jobs a month that would have gone to voicemail, what is that worth against the monthly? Let them do that division out loud.

This structure matters because of cycle length. No build fee means no five-figure number in the room, which means no financing conversation and no three-week delay while they think about it. We go deeper on that trade-off in should you charge a build fee and on setting retail in how much should you charge for an AI buildout.

Who should you not sell this to?

This is the section most vendors skip. Four kinds of plumbing company where you should pass, or sell something much smaller.

The one-truck owner-operator. If the owner answers the phone himself and runs twelve jobs a week, the missed-call problem is real but small, and your monthly is a meaningful share of his overhead. Sell him a single service if he insists, or tell him to call you when he hires his second tech. He will remember that you said it.

Shops already inside a private-equity platform. PHCC's scan names NexCore and PremiStar among the platforms acquiring regional contractors. If a shop has been rolled up, the phone stack is a corporate decision and the local GM cannot buy from you. Qualify for ownership structure early rather than finding out in week three.

Commercial and industrial contractors. Their inbound is purchase orders, bid invitations and property managers with account numbers, not homeowners in a panic. An agent tuned for residential emergency triage is the wrong product. It can be built for them. It is a different build and a longer sale.

Anyone who wants the AI hidden. Owners will ask whether callers can tell. The honest answer is that a good agent is convincing on a short call, and that you should not design around people failing to notice. Where disclosure practice stands is covered in do you have to tell callers it's an AI. A prospect whose main interest is deception makes a bad account and a worse reference.

Two limits on our side, plainly. Your $49 plan publishes one demo at a time and caps single calls at five minutes, which is fine for demos and is not a production phone system. And the demo's booking actions are sandboxed, so nothing touches a live calendar until the real build happens in fulfillment. What that build involves after signature is laid out in what happens after your client signs.

What changed recently that helps this sale

Two things, and they pull in the same direction.

The first is that the category stopped being exotic. The Federal Reserve's April 2026 note on monitoring AI adoption puts firm-level adoption at about 18 percent as of year-end 2025 in the Census Business Trends and Outlook Survey, with over 20 percent of firms expecting to use AI in the first half of 2026. Quote it carefully. Census broadened the question in November 2025, so that 18 percent is not comparable to the roughly 10 percent the older series showed. The note also finds adoption tracks firm size while the smallest firms adopt more than size alone would predict, which is the cohort your buyer sits in.

The second is that their customers changed. The same Pew survey that found 63 percent uneasy about the pace also found about half of US adults now use an AI chatbot. The homeowner calling your client at 9:40pm has talked to a bot this month. That does not mean they want to talk to one now. It does mean the interaction is familiar, and familiarity is why these calls stopped ending in hangups the way they did two years ago.

The version of this argument you should not make is that AI is inevitable and they are falling behind. Plumbers have heard inevitability pitches about every piece of software sold to them since 2015. Make the argument about Saturday night.

Your first two weeks selling into plumbing

Week one, build one demo and keep it. Pick a local plumbing company with reviews mentioning a missed call or a slow callback, put their name on the demo, and call them. Build it before the call, because "let me put something together for you" is where these deals die.

Then use that demo twenty times. The plumbing pack does not change between prospects. Only the business name and the audited domain do. Reps who treat each demo as a custom project make four calls a week. Reps who treat it as a template with two variables make forty.

Week two, go where they already are. Supply houses in the morning. Local contractor groups. The plumbing Facebook groups where owners complain about their answering service by name. That last one is the shortest path to a qualified conversation you will find, and it costs nothing.

If you want the same treatment for an adjacent trade, the HVAC playbook covers a buyer with a seasonal call spike instead of an emergency one, and most reps end up selling both.

FAQ

Do plumbing customers accept an AI answering the phone?

More than they did, and tolerance is highest in the moment that matters. Somebody with water spreading across a floor wants a fast answer and a callback time. They are not evaluating the voice. The harder audience is the scheduled-work caller with no urgency, which is why the agent should book those fast and stop talking.

What size plumbing company is the right target?

Three to six trucks. Big enough that calls get missed and the owner is not the dispatcher, small enough that they can decide on the call. Below that the payback is thin. Above roughly fifteen trucks you are usually dealing with an existing phone stack or a corporate parent.

Can the agent quote plumbing prices?

It should quote bands and book a diagnostic, not commit to a firm number on work that needs an inspection. The vertical pack carries price bands for the trade, and during onboarding your client supplies their own. Any agent that gives a hard quote on a job it cannot see is a liability you are handing your client.

How long before a plumbing client's system is live?

Your demo is live in about ten minutes, most of it choosing a subdomain and uploading a logo. The client build runs through the fulfillment platform after signature, and the timeline depends on how fast they return their hours, services, price bands and calendar access. Onboarding help chases that so you are not the bottleneck.

Do I need to buy a phone number or an API key?

No. Voice and chat run on our infrastructure and demo minutes are included in the subscription. Nothing to configure before your first call, which matters when an owner offers you a twenty-minute window on Thursday.

The next step

Open the live demo on our homepage, switch the trade, and call the receptionist as if you were a homeowner with a burst pipe. Note the three questions it asks before it does anything else. Those three questions are your pitch. Then start your workspace, build one plumbing demo with a real local company's name on it, and call them Monday morning before the trucks roll.