What 62% of missed calls actually cost a 6-truck shop — and the AI receptionists worth comparing

Most service-trade shops are quietly losing five figures a quarter to one phone setting. 62% of inbound calls in home services go unanswered. Of callers who can't reach a business on the first try, 85% never call back.

Source: LeadTruffle 2026 home-services research, aggregated across HVAC, plumbing, electrical, and roofing inbound traffic.

This isn't an "industry-wide average we should improve." This is the operating reality of most small service-trade shops. If you've never measured your own missed-call rate, the right assumption is that you're somewhere in that 62% band — until proven otherwise.

The math for a 6-truck HVAC shop

Take a typical 6-truck HVAC operator. Rough inputs:

• Inbound calls per week: ~50 (new leads, repeat customers, callbacks, service questions)
• Average ticket on closed work: $410 (ServiceTitan 2025 Industry Report)
• Close rate on captured leads: ~30% (industry-typical)
• Missed-call rate: 62%

Run the math:

Three weeks of that compounds to $9,840 in revenue the shop never sees. Annualized: roughly $170,000 in unrealized pipeline.

This is a modeled scenario, not a specific shop. But if your numbers are within 30% of the inputs above, your shop is leaking five-figure quarterly revenue to a problem most owners don't even measure.

Why this happens

Three structural reasons missed-call rates run this high in service trades:

1. The owner answers the phone for the first three years, then gets pulled into the field on bigger jobs and drops out of inbound coverage. The shop never replaces that coverage with infrastructure.
2. The legacy fix — voicemail and a callback — is broken. Modern customers (especially under-50 homeowners who increasingly drive HVAC and plumbing budget decisions) treat voicemail as failure. They call the next contractor.
3. After-hours coverage is treated as optional. ~25–35% of service-trade calls hit outside 9–5 hours, and emergency-service calls (water heater failure, AC out in summer) are the highest-ticket, highest-urgency leads. Voicemail loses these almost universally.

Three fixes, ranked by ROI

1. Voicemail-to-text via your carrier

Cost: $0/mo · Setup: ~30 min

Almost every cell carrier and VoIP provider offers automated voicemail transcription at zero additional cost. Verizon, AT&T, T-Mobile, RingCentral, Grasshopper, and OpenPhone all include it.

What it solves: you read the message faster than you'd listen to it. You get the callback context (their address, what's broken) without the 90-second voicemail listen.

What it doesn't solve: the lead still went to voicemail. Most people who hit voicemail in 2026 hang up.

ROI: Free. Marginal improvement.

2. Ring-deep + mobile forwarding

Cost: $0–$10/mo · Setup: ~15 min

Configure your business line to ring 4 times before forwarding to your mobile, instead of dropping to voicemail at ring 2. Most VoIP providers have this in settings under "ring routing."

What it solves: catches calls that hit during a 5-minute meeting or a phone-out-of-pocket moment. Most shops have this set wrong by default.

What it doesn't solve: after-hours coverage. Calls outside business hours still drop.

ROI: Marginal. Free. Takes 15 minutes.

3. Entry-tier AI receptionist

Cost: ~$25/mo · Setup: 1–3 hours

This is the actual fix. AI voice agents that answer your business line 24/7, ask the right intake questions, schedule the appointment in your CRM if you have one, and either text you the lead or transfer hot calls to your mobile.

The category went from "experimental" to "production-ready" in 2024–2025. Most cost less than your current phone bill.

The AI receptionist landscape — April 2026

Tuesday Teardown #2 walks the mechanism — what an AI answering layer actually does during an inbound call — and the math at a 6-truck shop modeled against the cost above. We deliberately don't crown a vendor; the right one for your shop depends on your CRM, call volume, and lead mix, which is what the Operator's Audit answers. For now, here's the lay of the land in the category.

• Trillet ($49/mo, 150 minutes included) — voice + SMS + WhatsApp multi-channel. Best for small shops that want emergency-text alerts when a call comes in.
• Marlie.ai ($0.19/min pay-as-you-go) — native integrations with Housecall Pro and Jobber, so the bot can write directly into your CRM. Best for shops on those platforms.
• Smith.ai ($95+/mo) — hybrid AI + live agent. AI handles the 80%, humans handle the edge cases. Best for shops that don't want to fully give up the human voice.
• LeadTruffle ($229–$629/mo) — integrates with Thumbtack, Angi, and other lead aggregators. Best for shops that buy from those platforms and want to keep response time under the threshold for lead-quality scoring.
• Upfirst ($24.95/mo entry tier) — cheapest of the named players. Solo-operator focus.
• Avoca (enterprise pricing) — Kleiner Perkins-backed, $1B valuation as of April 2026. Too expensive for shops under 30 trucks but signals where the category is heading.

ROI math on the AI receptionist tier

Take the same 6-truck HVAC shop above. Current state: 31 missed calls/week, 8 jobs/week left on the table, $3,280/week lost pipeline.

If the AI catches 70% of missed calls and converts at the same 30% close rate as the rest of the funnel:

This is the kind of math that most shops don't run.

What not to do

• Don't switch CRMs to "fix" missed calls. ServiceTitan / Housecall Pro / Jobber all have AI receptionist add-ons or partner integrations now, but the receptionist itself is the lever, not the CRM. Don't conflate the two.
• Don't hire a part-time human receptionist as the first move. Average trade-shop hourly cost: $35–$55/hr loaded. Even 10 hours/week of coverage is $1,400–$2,200/mo — 30–40× the AI cost — and humans don't cover after-hours without paying premium.
• Don't go straight to the most expensive option. The category has rapidly compressed in the last 12 months. The $25/mo entry tier covers most of the use case for a 5–10 employee shop.

How to actually set this up

Concrete steps. Most service-trade operators can have an AI receptionist live by the end of a Saturday afternoon. Here's the playbook.

Step 1 — Pick the right tool for your shop (15 min)

Your CRM and shop size are the primary filters.

• Housecall Pro or Jobber: Marlie.ai (~$0.19/min). Native CRM integrations — the bot writes new leads directly into your existing pipeline. No middleware.
• ServiceTitan: ServiceTitan's own Voice AI add-on or Avoca's enterprise integration. Stay inside the ServiceTitan ecosystem so jobs flow through your existing dispatch view.
• 2–5 trucks, no major CRM: Upfirst ($24.95/mo) or Trillet ($49/mo). Either ships with a workable default intake flow you can adapt in 30 minutes.
• Heavy on Thumbtack/Angi leads: LeadTruffle ($229+/mo). Their integration with lead aggregators keeps response time under the auto-deprioritization threshold those platforms enforce.
• Default if you're unsure: Trillet at $49/mo. Multi-channel (voice + SMS), small-shop focus, comes with a 30-day trial.

Step 2 — Sign up and configure (~45 min)

• Create the account.
• Provide your business name, hours, services offered, and service area zip codes.
• Upload your existing FAQ if you have one (most tools accept a Word doc or pasted text).
• The tool gives you a forwarding number. Forward your business line to it during the after-hours window only as a starting point.
• Walk through the tool's onboarding call (most are 20–30 minutes).

Step 3 — Connect to your CRM (varies)

• Housecall Pro / Jobber: native integrations on Marlie.ai, Trillet, and Smith.ai. Click-through OAuth — no IT required.
• ServiceTitan: works with Avoca, ServiceTitan Voice AI directly, or via Zapier with Smith.ai.
• No CRM: the tool will text you new leads directly. Add to your task system manually until you adopt a CRM.

Step 4 — Tune the intake script (~30 min)

Default scripts are generic. The 30 minutes you spend customizing pay back in close-rate. Customize for:

• Trade-specific qualifying questions. HVAC: "Is the unit completely off, or just not cooling?" Plumbing: "Is water actively running or contained?" Roofing: "Active leak or insurance claim?"
• Service area boundaries. Which zip codes you cover. The bot should politely decline anything outside the radius.
• Pricing transparency. Tell the bot your trip charge and minimum diagnostic fee. Customers who hear the number stay; the ones who don't, you weren't going to close anyway.
• After-hours emergency criteria. Which calls should ring your mobile (water heater leak, no AC in summer, exposed wiring) vs. wait until tomorrow (general questions, scheduling).

Step 5 — Test with 5 sample calls

Have a friend, family member, or one of your techs run five scenarios through the new line:

1. Standard service request ("My AC isn't cooling")
2. Emergency ("Water heater just burst")
3. Quote request ("How much for a new system?")
4. Existing customer ("My last service was three months ago")
5. Wrong-area caller (a zip outside your service radius)

Listen to the recordings. Tune the script for any gaps. Most tools let you do this in 10 minutes.

Step 6 — Cutover plan (3 weeks)

• Week 1. After-hours only. Forward business line to the AI from 5pm to 8am and weekends. Calls during business hours still ring you directly.
• Week 2. After-hours + ring-deep failover. Calls during business hours ring you four times; if you don't pick up, they roll to AI instead of voicemail.
• Week 3+. Decide based on data. If the AI is converting captured calls at parity with your manual answer rate, route 100% of overflow to it.

Step 7 — Tune over 30 days

Pull the AI's call recordings weekly. Tune for:

• Repeated questions the AI fumbles or escalates unnecessarily
• Booking-flow drop-offs (where in the script the customer hangs up)
• Customer feedback on the intake experience (text them after, ask 1 question)

Most tools let you A/B test scripts. Do it. By day 30 you'll have a script that converts better than your live answer ever did.

The 90-day play

If you're starting from "we have voicemail":

• Week 1. Voicemail-to-text on. Ring-deep configured. Free fixes done.
• Week 2. Pick one AI receptionist from the $25–$95/mo tier. Sign up, do the configuration call (most have a 30-minute onboarding). Forward your business line to it during after-hours only as a starting test.
• Week 3. Run an A/B — half of inbound goes to the AI, half goes to your existing flow. Measure: capture rate, conversion to booked appointment, customer feedback.
• Week 4. Decide. If the AI is catching 60%+ of formerly-missed calls and converting at parity with your normal flow, route 100% of overflow to it.
• Weeks 5–12. Tune. Adjust the intake script. Add the integrations you need (CRM, scheduling, payment). Iterate on the after-hours decision flow.

The shop that runs this 90-day playbook ships ~$30,000 of recovered pipeline by week 12. The math is one-sided.

Want this run on your specific shop?

The numbers above are modeled on industry averages. Your shop's actual numbers will differ — call volume, ticket size, current capture rate, after-hours mix, your existing CRM. The Operator's Audit takes a 30-question intake on your shop and returns a 12-page roadmap with the recommendation, the integration steps, and the dollar-modeled impact for your specific size and trade.

The first three audits we run go to the first three readers to reply — total, not per reader. They're free in exchange for an honest written take on whether the work was useful. Public price after the first three is $297. See the Audit page for the full scope, or reply to the next Tuesday Teardown to claim one of the three slots.

Sources

• Inbound missed-call rate (62%) and call-back rate (15%): LeadTruffle 2026 home-services research
• HVAC average ticket ($410): ServiceTitan 2025 Industry Report
• AI receptionist pricing: vendor public pricing pages, verified April 2026
• Avoca $1B valuation: Fortune, April 27, 2026
• After-hours call mix (~25–35%): industry estimates, modeled
• Trade-shop loaded hourly cost ($35–$55/hr): industry estimates, modeled