Your dispatcher picks up the phone. “Hi, I’m just calling to see when the technician will be there.” She glances at the board, checks the last GPS ping, does the mental math on traffic, and gives her best guess. Three minutes later, another call. Same question, different customer. By lunch she’s answered this question eleven times.
Every one of those calls is dead time. She’s not booking a new job, she’s not following up on an estimate, she’s not ordering parts. She’s reading aloud from a screen the customer could see themselves if the information were in their pocket.
Trades businesses doing $2M to $15M in revenue lose 15 to 25 hours per week to “where’s my tech?” calls. That’s half a full-time CSR salary spent on status updates. The fix isn’t hiring another body to answer the same question faster. The fix is making the question disappear.
Why customers call to check arrival time
They call because they don’t know. The appointment window you gave them was 8 a.m. to noon. It’s 10:30. They have a meeting at 1 p.m. and they’re wondering if they should reschedule it. Or they’re sitting in the driveway of a rental property and the tenant is texting them every twenty minutes.
You can’t blame them. A four-hour window is a guess, not a promise. Jobs run long. Traffic happens. A parts run adds forty minutes. Your tech finishes early and you slide the next job forward. All of that is normal dispatch work, but the customer doesn’t see it. They see a window that’s closing and no update.
So they call. And your CSR stops what she’s doing, pulls up the dispatch board, finds the job, checks the tech’s location, estimates drive time, and relays it. If the customer calls back an hour later because the estimate was off, she does it again.
The problem isn’t the customer. The problem is that the information lives in your dispatch system and dies there. It never makes it to the person who needs it most.
What real-time arrival notifications actually do
Real-time arrival notifications push the information to the customer before they think to ask. When your tech finishes the job before theirs, the system recalculates the ETA and sends an update. When he’s fifteen minutes out, they get a text with a tracking link. When he pulls into the driveway, they get a “your tech has arrived” message.
The customer opens the link and sees a map with a dot moving toward their house. They see the tech’s name, his photo, and the updated arrival time. They don’t call because they already know.
One HVAC company in our network rolled out arrival notifications in March. Their CSR tracked inbound calls for two weeks before and two weeks after. “Where’s my tech?” calls dropped by 78%. The calls that did come in were about scope changes, access issues, and new work, things that actually required a human conversation.
The CSR didn’t get less busy. She got busy with different work. She started calling back the estimates that had gone cold. She followed up with customers who’d asked for a quote three weeks ago and never heard back. Revenue per week went up because she had time to do the work that closes jobs instead of the work that just keeps people calm.
How AI handles the entire loop
Manual arrival notifications don’t scale. If your CSR has to send each text herself, you’ve just traded phone time for texting time. The system has to run itself.
An AI-powered dispatch agent watches your dispatch board in real time. When a job status changes, when a tech clocks out, when GPS shows movement toward the next address, the agent updates the ETA and sends the message. No one on your team touches it.
The 24/7 Dispatch Voice Agent we build for trades businesses does more than answer the phone. It qualifies the job, books the slot, and immediately sends the customer a confirmation text with a tracking link. The link goes live the morning of the appointment. The customer can check it whenever they want. If they call anyway, the voice agent answers and tells them exactly where the tech is and when he’ll arrive. Your CSR never picks up the phone.
The Estimate Follow-Up Agent layers on top of this. After the job is done, it tracks whether an estimate was sent. If the customer hasn’t responded in two days, the agent sends a follow-up message. If they still haven’t responded by day five, it sends another, this time with a line about scheduling filling up. On day fourteen, it makes one last push. Fifteen to twenty-five percent of stale estimates convert when you follow up consistently. Most trades businesses don’t follow up at all because there’s no one with the time to do it.
The Review and Reactivation Agent closes the loop. The day after the job, it asks the customer for a review. If they’re happy, it sends a link. If they’re not, it flags the job for your ops manager to call. Six months later, if the job was seasonal work like an AC tune-up or a furnace check, it sends a reactivation message. Repeat customers are 40% cheaper to win than new ones, but only if you remember to ask.
These three agents work together. The first one eliminates the inbound question. The second one turns completed work into closed revenue. The third one turns closed revenue into repeat revenue. None of it requires your CSR to remember, to check a list, or to carve out time.
What it looks like in practice
A customer books a water heater replacement on Tuesday for Thursday morning. The voice agent answers the call, confirms the address, asks a few qualifying questions (gas or electric, access to the unit, age of the home), and books the slot. The customer gets a confirmation text with a link.
Thursday morning at 7 a.m., the system sends a reminder: “Your technician will arrive between 8 and 11 this morning. Track his arrival here.” The customer clicks the link and sees a map. Your tech finishes his first job at 9:15 instead of 10. The system recalculates and sends an update: “Your technician is running ahead of schedule and will arrive around 9:45.”
At 9:30, the customer gets another message: “Your technician is 15 minutes away.” At 9:44, the tech pulls up and the system sends a final message: “Your technician has arrived.”
The customer never called. Your CSR never answered. The tech showed up, did the work, and left. That afternoon, the estimate follow-up agent sends a message: “We sent your quote this morning. Let us know if you have any questions or if you’d like to get on the schedule.” The customer replies yes, and the agent hands the conversation to your CSR to finalize.
Two days later, the review agent sends a message: “How did we do? We’d love to hear your feedback.” The customer leaves a five-star review. Six months later, the reactivation agent sends a note: “It’s been six months since we replaced your water heater. If you’re due for a furnace check before winter, we have a few slots open in October.”
That’s the full loop. Booking, arrival, follow-up, review, reactivation. The only human touch was your CSR finalizing the schedule after the customer said yes to the estimate. Everything else ran on its own.
If you want a structured way to think through what calls you’re missing after hours and how much revenue is walking away, we built a worksheet that maps it out. The After-Hours Call Recovery Plan walks you through peak call windows, average job size, and close rate so you can calculate the dollar cost of every missed ring. It’s a fifteen-minute exercise that usually surfaces $3K to $8K per month in recoverable revenue.
What you’re actually buying back
The obvious win is time. If your CSR spends twenty hours a week answering “where’s my tech?” calls, you’re buying back half her week. That’s ten hours she can spend calling estimates, booking follow-up work, or handling the calls that actually generate revenue.
The less obvious win is customer experience. A customer who has to call you to find out when you’re coming is already halfway to annoyed. A customer who gets proactive updates feels taken care of. They’re more likely to leave a review, more likely to call you next time, and more likely to refer you to a neighbor.
The third win is dispatch flexibility. When customers can see real-time updates, you can move jobs around without triggering a flood of phone calls. If a furnace install runs two hours over and you need to push the next three jobs back, the system sends updated ETAs automatically. Your CSR doesn’t field a dozen “I thought you’d be here by now” calls. The customer sees the new time and adjusts.
One electrical contractor we work with runs a tight schedule in new construction. Builders expect them on-site at a specific time because other trades are waiting. Before arrival notifications, the site super would call every morning to confirm the time. After, the system sent an update when the crew left the previous job and another when they were ten minutes out. The supers stopped calling. The relationship got easier because the information flowed without anyone having to ask for it.
How to know if this is worth building
If your CSR answers more than five “where’s my tech?” calls per day, you’re losing time. If she answers more than ten, you’re losing money. At fifteen calls per day, you’re paying someone $18 to $25 per hour to read a dispatch board out loud. That’s $6K to $9K per year per CSR just on arrival-time questions.
The cost to build and run an AI agent that handles this automatically is a fraction of that. Most trades businesses we work with see payback in under ninety days, and that’s before you count the revenue lift from better follow-up and reactivation.
The businesses that get the most value from this are the ones where the phone rings constantly and the person answering it is already underwater. If your CSR is also your dispatcher, your bookkeeper, or your owner’s spouse, every minute she spends answering “where’s my tech?” is a minute she’s not spending on something that moves the business forward.
The second group that benefits are businesses with tight appointment windows or high-value jobs. If you’re doing $3K to $8K service calls and the customer is taking time off work to be home, a proactive arrival notification is the difference between a smooth experience and a tense one. The customer who feels informed is the customer who pays on time, leaves a review, and calls you next year.
What an Omni Audit uncovers
We built the AI audit for trades businesses to map where your time and money are leaking before we write a single line of code. It’s a sixty-minute working session. You walk me through a typical week: how calls come in, how dispatch works, how estimates go out, how follow-up happens (or doesn’t).
I’ll ask about your busiest call windows, your average job size, your close rate on estimates, and how many calls you think you’re missing after hours. We’ll look at your dispatch tool, your CRM if you have one, and your phone system. By the end, you’ll have three things: a process map that shows where the manual work is happening, a leakage estimate that puts a dollar figure on it, and a ranked list of agents we’d build first.
Most trades businesses find $50K to $200K per year in recoverable revenue. Half of that is missed calls and stale estimates. The other half is repeat work that never got reactivated and reviews that never got asked for. The audit doesn’t fix any of it, but it shows you exactly what’s fixable and what the return looks like if you fix it.
What happens after the audit
If the numbers make sense, we build the first agent. For most trades businesses, that’s the 24/7 dispatch voice agent. It goes live in two to four weeks. We connect it to your phone system and your dispatch tool. It answers every call, qualifies the job, books the slot, and sends the confirmation with a tracking link.
You’ll see the impact in the first week. Inbound call volume drops. Your CSR has time to breathe. Customers stop calling to check arrival time because they already know.
Once the first agent is running, we layer in the second. Usually that’s the estimate follow-up agent. It watches every estimate that goes out and follows up automatically on day two, day five, and day fourteen. You’ll start closing estimates you would have lost to silence.
The third agent is typically the review and reactivation agent. It asks for reviews the day after the job and reactivates customers at the right interval. Your review count goes up, your repeat business goes up, and your cost to acquire a customer goes down because more of your revenue is coming from people who already know you.
We don’t build all three at once. We build one, let it run, measure the result, and then build the next. You’re never paying for something that isn’t working yet. You’re paying for the agent that’s live and generating value today.
If you want to see what other trades businesses are building with Omni, the AI operations hub has examples and case studies. If you want to understand how the voice agent works in more detail, the voice agent page walks through the architecture and the integration points.
The cost of waiting
Every week you wait is another week your CSR spends twenty hours answering “where’s my tech?” calls. If she’s making $50K per year, that’s $480 per week in labor cost just to read a dispatch board out loud. Over a year, that’s $25K.
The revenue cost is harder to see but bigger. Every estimate that doesn’t get followed up is a coin flip you’re not taking. Every customer who doesn’t get asked for a review is a missed signal to the next customer. Every repeat customer who doesn’t get reactivated is a new customer you have to spend $400 to $900 to acquire instead.
Most trades businesses we work with find that the cost of waiting one quarter is equal to the cost of building the entire system. The businesses that move fast see payback in sixty to ninety days. The businesses that wait see the same problems six months later and wonder why they didn’t start sooner.
Want the practical version of this? The free Working With Claude field guide covers the full Claude ecosystem, Claude Code, and how to roll it out across a real business. Download it here.