You’re on a job site pulling wire or under a sink when your phone rings. You let it go to voicemail because you’re elbow-deep in the work. When you check later, there’s no message. That call was probably worth $800 to $2,500, and it’s gone. The homeowner already rang the next electrician or plumber on Google.

This happens 10 to 20 times a week in a typical trades business doing $2M to $8M. The math is brutal. If half those calls don’t leave a message and you miss even 30 percent of the ones that do, you’re bleeding $50K to $200K a year in jobs that never make it onto the schedule.

The same pattern plays out on the back end. You send an estimate for a furnace replacement or a panel upgrade. The customer says they’ll think about it. You mean to follow up in a few days, but three more emergencies land and a week goes by. Then two weeks. The estimate goes cold. Industry data shows that 15 to 25 percent of stale estimates convert when someone follows up with the right message at the right time, but most trades businesses never make that second call.

AI agents built for trades work can close both leaks. A voice agent answers every inbound call 24/7, qualifies the job, books the slot, and texts a confirmation. An ops agent tracks every estimate you send and follows up on day two, day five, and day 14 until the customer says yes or no. A third agent asks every happy customer for a review the day after you finish and reactivates past customers when it’s time for their next service.

This isn’t theory. We’ve built these agents for HVAC contractors, electricians, plumbers, and roofers. The work is specific, the ROI is measurable, and the setup starts with a 60-minute audit that maps exactly where your revenue is leaking and which agent closes it first.

The Real Cost of Missed Calls in a Trades Business

Most owners know they’re missing calls. What they underestimate is how many and what each one costs.

A plumbing business doing $3M a year typically fields 40 to 60 inbound calls a week. Half are existing customers with questions or scheduling changes. The other half are new leads: a water heater that died, a clogged main line, a bathroom remodel estimate. If you’re on the tools or dispatching a crew, you’re not picking up. Your voicemail greeting says you’ll call back, but 40 to 50 percent of callers don’t leave a message. They move to the next search result.

Let’s say you miss 12 new-lead calls a week and half don’t leave a message. That’s six jobs that never enter your pipeline. If your average service call is $650 and your close rate on calls you do answer is 60 percent, you’re losing roughly four jobs a week. Over a year, that’s 200 jobs and $130K in top-line revenue. For a business running 20 percent net margin, that’s $26K straight to the bottom line that you never saw.

The problem compounds when your admin or spouse is answering calls between other tasks. They can take a message, but they can’t qualify the urgency, check the schedule, or book the slot. The customer hears “someone will call you back” and starts shopping. Speed matters. A lead that gets a callback in two hours converts at half the rate of one that books on the first call.

Emergency calls are worse. A homeowner with no heat in January or a sparking panel isn’t leaving three voicemails. They’re calling until someone picks up. If you’re the second or third call, you’ve lost a $1,200 furnace repair or a $2,800 panel replacement to the guy who answered first.

This is the gap a 24/7 Dispatch Voice Agent fills. It picks up every call, asks the right questions (What’s the issue? Is this an emergency? What’s your address?), checks your real-time schedule, and books the job or routes it to your on-call tech. The customer gets a confirmation text with the time window and your tech’s name. You get a ticket in your dispatch system with all the details. No voicemail tag, no lost lead, no second-guessing whether the call was worth $500 or $5,000.

One HVAC contractor we work with had been losing roughly 15 calls a week to voicemail during peak summer months. After deploying the voice agent, inbound booking rate jumped from 52 percent to 81 percent. That’s an extra $140K in booked revenue in 90 days, most of it work that would have walked.

Why Estimates Die and How AI Brings Them Back

You spend 90 minutes walking a homeowner through a whole-house generator install or a roof replacement. You write up a detailed estimate, email it that night, and wait. The customer says they need to talk it over or get another quote. You make a note to follow up, but tomorrow brings two no-heat calls and a supply house run. A week slides by. Then two. The estimate is dead.

This is the silent killer in trades businesses. Most owners track their close rate on estimates they chase, but they don’t track how many estimates never get a second touch. When we run an audit for a trades business, we typically find 30 to 50 open estimates sitting in the CRM or filed in a folder with no follow-up activity. At an average estimate value of $3,200, that’s $96K to $160K in stalled pipeline.

The fix isn’t complicated, it’s just manual. Someone has to review every estimate, decide when to follow up, write a message that doesn’t sound pushy, send it, log the response, and repeat. For a business sending 15 to 25 estimates a month, that’s 10 hours of admin work no one has time to do.

An Estimate Follow-Up Agent does it automatically. When you send an estimate (or when it’s logged in your system), the agent creates a follow-up sequence. Day two: a friendly check-in asking if they have questions. Day five: a reminder that the quote is good for 30 days and a note about your next availability. Day fourteen: a last-chance message with a small urgency hook, like mentioning that material costs are rising or that your schedule is filling up for the season.

The messages are written in your voice, tuned to the trade and the job type. A $1,200 water heater replacement gets a shorter, simpler sequence than a $28,000 HVAC system. The agent tracks opens, clicks, and replies. If the customer responds, it flags you or your admin to take over. If they go quiet, it logs the outcome and moves on.

We see conversion rates of 18 to 26 percent on estimates that were sitting cold for more than a week. That’s incremental revenue most businesses assume is lost. For a contractor doing $5M a year with 200 estimates in play, recovering even 20 percent of the stale ones adds $120K to the top line with no additional marketing spend.

One electrical contractor told us he used to write “follow up Friday” on a sticky note every week and still only managed to chase half his quotes. After deploying the agent, every estimate gets three touches automatically. His close rate on estimates older than five days went from 11 percent to 23 percent. He didn’t change his pricing or his pitch. He just stopped letting quotes die in silence.

If you want a structured way to think through which calls and estimates are leaking in your business, we built a worksheet that walks you through the math. Grab the After-Hours Call Recovery Plan for Trades and spend 20 minutes mapping your real numbers. It’s a useful forcing function even if you’re not ready to deploy an agent yet.

The Follow-Up Work No One Has Time to Do

Closing the estimate is only half the battle. Once the job is done, there are three follow-up tasks that drive long-term revenue but almost never happen consistently: asking for a review, checking in to make sure the customer is happy, and reactivating them when it’s time for the next service.

Every trades business owner knows reviews matter. A plumber with 80 Google reviews and a 4.8-star average gets three times the inbound call volume of a competitor with 15 reviews and no recent activity. But asking for a review feels awkward, and timing matters. Ask too soon and the customer hasn’t formed an opinion. Ask too late and they’ve moved on.

The manual process looks like this: finish the job, send an invoice, make a note to text the customer tomorrow, forget, remember three days later, send a generic “we’d love a review” message that gets ignored. Repeat for every job. Most businesses get reviews from 5 to 10 percent of completed jobs, almost always the customers who were either thrilled or furious.

A Review and Reactivation Agent automates the whole cycle. The day after the job closes, it sends a message thanking the customer and asking how everything went. If the response is positive, it immediately follows up with a direct link to leave a review on Google, along with a short suggested template they can edit. If the response is neutral or negative, it flags you to call and make it right before the customer posts anything public.

The same agent tracks service intervals. If you installed a furnace, it knows to reach out in 11 months to schedule the first annual tune-up. If you did an electrical panel upgrade, it checks in at the one-year mark to see if the customer needs any additional circuits or outlets. These reactivation messages convert at 12 to 18 percent, which is higher than cold outreach and costs nothing.

One HVAC contractor running this agent went from 22 Google reviews to 140 in nine months. His inbound call volume doubled, and his cost per lead dropped by 40 percent because more customers were finding him organically. He didn’t change his service quality. He just asked every customer at the right moment.

The reactivation piece is even more valuable for businesses with recurring service work. A plumber doing annual backflow testing or a roofer offering gutter cleaning can build a predictable revenue base by reactivating past customers at the right interval. Most businesses lose 60 to 70 percent of past customers simply because no one reaches out when it’s time for the next service.

What an AI Agent Actually Does (and What It Doesn’t)

There’s a lot of noise in the AI space right now. Chatbots that can’t handle a real conversation. Tools that promise to “automate your business” but can’t integrate with your dispatch software. It’s worth being specific about what a well-built agent does and where a human still needs to be in the loop.

A 24/7 Dispatch Voice Agent answers the phone, understands natural speech, and can handle the most common inbound scenarios: booking a service call, triaging an emergency, taking a message for a callback, and answering basic questions about your hours and service area. It connects to your scheduling tool (ServiceTitan, Housecall Pro, Jobber, or whatever you use) and books directly into an open slot. It sends a confirmation text and logs the call details.

What it doesn’t do: give pricing over the phone for complex jobs, handle an angry customer who wants to argue about a bill, or make judgment calls about whether a job is safe to defer. Those get routed to you or your dispatcher with a flag. The agent is smart enough to know its limits.

An Estimate Follow-Up Agent monitors your CRM or estimate log, sends the right message at the right time, tracks responses, and flags hot leads for you to close. It doesn’t negotiate pricing or rewrite the estimate. It keeps the conversation alive until you can step in.

A Review and Reactivation Agent asks for feedback, routes happy customers to your review page, flags unhappy ones for you to call, and reactivates past customers when it’s time for their next service. It doesn’t handle complex service questions or try to upsell over text. It tees up the opportunity and hands it to you.

The goal isn’t to replace your dispatch process or your sales process. It’s to handle the repetitive, time-sensitive work that leaks revenue when it doesn’t happen fast enough or consistently enough. The agent works 24/7, never forgets a follow-up, and costs a fraction of a full-time admin.

We’ve written more about how these agents fit into the broader operational stack for trades businesses. If you want to see the full picture, take a look at the AI audit for trades businesses and the specific agents we build for plumbing, HVAC, electrical, and roofing companies.

How to Know Which Agent to Build First

Most trades businesses have all three leaks: missed calls, dead estimates, and no follow-up. The question is which one to close first.

The answer depends on where the biggest dollar value is leaking and where you have the least manual coverage today. If you’re missing 10 to 15 calls a week and you don’t have a dedicated dispatcher or answering service, the voice agent is the obvious first move. It pays for itself in the first month by converting calls that would have walked.

If you’re answering most calls but your estimate close rate is under 40 percent and you have 30-plus open quotes with no follow-up, the estimate agent is the better starting point. It turns stalled pipeline into closed revenue without adding headcount.

If your inbound volume is strong and your close rate is solid but you’re not getting reviews and your repeat business rate is under 20 percent, the review and reactivation agent is the highest-leverage play. It builds your organic lead flow and reactivates revenue you’ve already earned once.

We figure this out in the first 15 minutes of an Omni Audit. We look at your call volume, your CRM, and your estimate pipeline. We map the leakage, size the dollar impact, and recommend which agent to deploy first. Then we show you what the buildout looks like, how it integrates with your existing tools, and what the ROI timeline is. The whole conversation takes 60 minutes, and you walk out with three things: a leakage map, a prioritized agent roadmap, and a build quote.

No deck, no discovery call that turns into a sales pitch. Just a working session with someone who’s built these agents for dozens of trades businesses and knows what works. Book a 60-min Omni Audit and we’ll map your business in the first half hour.

What the Build Looks Like

Once you decide which agent to build, the process is faster than most owners expect. We’re not talking about a six-month IT project. A single agent typically goes live in two to four weeks.

Week one is integration and configuration. We connect the agent to your phone system, your scheduling tool, and your CRM. We write the scripts and decision trees based on how your business actually operates. If you route emergency calls differently than scheduled work, the agent learns that. If you have different pricing for commercial vs residential, it knows when to ask.

Week two is testing. We run the agent through 20 to 30 real-world scenarios, adjust the language and routing logic, and make sure it hands off cleanly to your team when it needs to. You listen to sample calls or review sample messages and tell us what sounds right and what doesn’t.

Week three is the soft launch. The agent goes live but we monitor every interaction for the first few days. If something breaks or sounds off, we fix it in real time. By the end of week three, the agent is handling 80 to 90 percent of interactions without any human intervention.

Week four is optimization. We review the data, see where customers are dropping off or getting confused, and tighten the scripts. We add edge cases we didn’t anticipate and tune the handoff triggers. By the end of the month, the agent is running autonomously and you’re reviewing a weekly summary instead of listening to every call.

After that, it’s maintenance and expansion. We add new capabilities as your business changes. If you launch a new service line, we update the agent’s knowledge base. If you hire a new tech and need to adjust the dispatch logic, we make the change in a day.

The cost structure is straightforward. You pay a build fee to get the agent live and a monthly subscription to keep it running. The build fee is typically $3K to $8K depending on how many integrations and how much custom logic you need. The monthly cost is $400 to $1,200 depending on call volume and complexity. For most trades businesses, the agent pays for itself in the first 30 to 60 days just from the calls and estimates it saves.

We’ve built agents for businesses using ServiceTitan, Housecall Pro, Jobber, FieldEdge, and a dozen other tools. We’ve integrated with Twilio, RingCentral, and plain old landlines. If you’re worried your tech stack is too old or too weird, it probably isn’t. We’ve made it work with dispatch boards running on Excel and CRMs held together with duct tape.

The technical side isn’t the hard part. The hard part is deciding to stop losing revenue to manual processes that don’t scale. Once you make that call, the build is just execution.

Why This Works Better Than Hiring

The obvious alternative to an AI agent is hiring someone to do the work. A full-time dispatcher or admin can answer calls, follow up on estimates, and ask for reviews. The problem is cost, consistency, and coverage.

A full-time dispatcher in most trades markets costs $40K to $55K a year once you factor in payroll taxes, benefits, and training time. That person works 40 hours a week, takes vacation, calls in sick, and quits when they find a better offer. If they’re good, you spend six months training them on your systems and your customer base. If they’re mediocre, you spend six months hoping they’ll improve before you start over.

An AI agent costs $5K to $15K a year, works 168 hours a week, never takes a day off, and improves over time instead of burning out. It doesn’t forget to follow up, doesn’t get distracted, and doesn’t leave for a competitor.

The consistency piece matters more than most owners realize. A human dispatcher has good days and bad days. They’re sharp in the morning and tired by 4 p.m. They’re patient with the first difficult customer and short with the fifth. An agent delivers the same quality on every interaction, whether it’s the first call on Monday morning or the last call on Friday night.

Coverage is the killer advantage. A trades business that wants to capture after-hours calls has three options: pay someone overtime to stay on call, route calls to an answering service that takes messages, or let them go to voicemail. All three options lose revenue. The on-call person is expensive and inconsistent. The answering service can’t book the job or check the schedule. Voicemail loses half the leads.

A voice agent answers every call, books the job, and confirms it in real time. It doesn’t matter if the call comes in at 6 a.m. or 11 p.m. The customer gets the same experience and you get the same revenue capture.

This isn’t an argument against hiring. If your business is growing and you need a full-time dispatcher to manage complex scheduling and customer relationships, hire one. But hire them to do the high-value work that requires judgment and relationship skills. Let the agent handle the repetitive, time-sensitive work that leaks when it’s done manually.

We’ve worked with trades businesses that deployed an agent and then hired a dispatcher six months later because the agent freed up enough revenue to justify the headcount. The dispatcher focuses on managing the schedule, handling escalations, and building customer relationships. The agent handles the inbound calls, the follow-ups, and the after-hours coverage. It’s a better division of labor and a better use of payroll dollars.

The Audit That Shows You Where to Start

We’ve talked about three agents, three leaks, and a dozen ways revenue walks out the door in a trades business. The next step is figuring out which leak is costing you the most and which agent closes it first.

That’s what the Omni Audit is for. It’s a 60-minute working session where we map your business, size the leakage, and build a prioritized roadmap. You walk out with three outputs: a leakage map that shows where revenue is slipping through, an agent roadmap that ranks which automations to build first, and a build quote that lays out cost and timeline.

We do this over Zoom. You share your screen and walk me through your dispatch process, your CRM, and your estimate pipeline. I ask questions about call volume, close rates, and follow-up cadence. We look at real data, not hypotheticals. If you’re missing 12 calls a week, we calculate what that costs. If you have 40 open estimates with no follow-up, we size the stalled pipeline.

By the end of the call, you know exactly where the biggest dollar leak is and what it would take to close it. No deck, no multi-phase proposal, no “we’ll get back to you in a week.” Just a clear map and a clear next step.

Most trades businesses that go through the audit deploy an agent within 30 days. Some decide the timing isn’t right and come back six months later when they have bandwidth to implement. Either way, the audit gives you a baseline so you know what you’re losing and what it would cost to fix it.

Book my Omni Audit and we’ll map your business in the first half hour. If you want to see what other trades businesses are building, see Omni for trades businesses and the specific agent types we deploy for plumbing, HVAC, electrical, and roofing companies.

The revenue is there. It’s just walking out the door while you’re on the tools or chasing the next emergency. An agent closes the leak. The audit shows you where to start.