Last spring, a residential solar installer was buying 850 leads a month and turning them into 110 booked site surveys. Ninety days later? Same lead budget, 176 site surveys on the calendar. A 60% jump, and not one new hire.
This is a real AI calling for solar case study from a regional installer in the US Southwest we worked with through 2025. They asked us not to name them, so the company stays anonymous and the numbers are theirs. They sell residential rooftop systems, run paid leads alongside web forms and referrals, and had five inside-sales reps handling intake, qualification, and booking.
The sales team wasn't the problem. The clock was. Their average speed to lead sat around 14 hours. Better than the 47-hour industry average, sure. But a homeowner who fills out a quote form at 9 PM doesn't wait 14 hours. They call the next installer on the list.
And 62% of their inquiries came in after the office closed or over the weekend. Those leads hit voicemail. By Monday they'd gone cold, or already booked a survey with a competitor who picked up.
Here's what they deployed, the before-and-after numbers, and the two situations where this still doesn't work.
Key Takeaways
- AI calling for solar cut the installer's speed to lead from 14 hours to under 60 seconds, day or night.
- Booked site surveys rose from 110 to 176 a month, a 60% increase on the same lead budget and team.
- Cost per booked site survey fell from $185 to $72 after the AI took over first touch and qualification.
- Two-touch SMS and voice reminders dropped the no-show rate from 30% to 18% with no rep effort.
- TopCalls runs first touch and qualification at $0.35 a minute, all-inclusive, responding in under 500 milliseconds.
1. The Challenge: 850 Solar Leads, Only 110 Site Surveys
Run the funnel backward. 850 leads in, 110 site surveys booked. That's a 13% lead-to-survey rate, and surveys are where the money gets made. No site visit, no quote, no install.
Two things were eating the other 87%. The first was leads going dark before anyone reached them. Roughly 1 in 4 never got a live conversation. A rep would dial twice, leave a voicemail, and move on to the next sheet.
When we pulled their CRM data, the leaks were everywhere:
Slow first touch: a 14-hour average response, against research showing leads contacted within five minutes are 21x more likely to convert than ones reached at 30 minutes.
After-hours blackout: 62% of inquiries arrived outside 9-to-5. Nobody answered, so those homeowners called somewhere else.
Junk in the pipeline: around 65% of purchased leads were never going to buy: renters, heavy tree shading, north-facing roofs, sub-650 credit. Reps still burned minutes qualifying each one.
No-shows: about 30% of booked surveys ghosted. Manual reminders were hit or miss, usually a single text the day before.
2. Why Speed to Lead Breaks Most Solar Sales Teams
Speed to lead is the metric that predicts whether a solar inquiry ever becomes a deal. And the decay is brutal: a lead loses roughly 80% of its value within five minutes of the inquiry going unanswered. Five minutes. Not five hours.
A human team can't win that race across 850 leads a month. You can't dial at midnight. You can't make 40 calls at once when a TV spot dumps 200 leads in an hour. Reps take lunch, they take days off, and call quality slides somewhere around dial number 200. None of that is a coaching failure. It's just what people are.
It's the same trap that hits the rest of home services. We broke down the pattern in our guide to AI appointment setting for home services like HVAC, solar, and roofing, and the speed-to-lead math is identical across all three.
3. AI Calling for Solar: What Solar Appointment Setting Actually Looks Like
So what does AI calling for solar actually do on a live lead? The short version: it answers in under a minute, every time, day or night.
The moment a new lead lands in the CRM, an AI voice agent dials out. It greets the homeowner by name, confirms they asked about solar, and walks through a qualification script that never varies. No rep had to be free. No lead waited overnight.
These aren't recordings or rigid phone menus. The AI voice agents hold a real back-and-forth, handle interruptions, and respond in under 500 milliseconds, so the homeowner feels like they're talking to a person, not a bot reading a card.
4. The Solution: What We Deployed in Two Weeks
Setup ran two weeks, from kickoff to live campaigns. Here's the stack we configured for them:
Instant outbound on every lead: the AI dials each new inquiry within 60 seconds, around the clock, weekends included. Busy and unanswered numbers get re-dialed automatically on a smart retry schedule.
A fixed qualification script: homeownership, roof age and type, shading, average monthly utility bill, credit band, and buying timeline. We mapped it to a consistent solar lead qualification flow so every homeowner got the same questions in the same order.
Direct calendar booking: qualified homeowners book a site survey right on the call. The AI handles solar appointment setting straight into the rep's live calendar, timezone-aware, no double bookings.
Automated reminders: an SMS and a short voice reminder 24 hours and 2 hours before the survey, part of a multi-touch follow-up sequence that ran without anyone watching it.
CRM sync and warm handoff: every call logged to HubSpot through our CRM integrations, with qualified homeowners routed to a human consultant for the visit itself.
Curious what this looks like at your lead volume? Drop your numbers into the ROI calculator and see your cost per booked appointment before you talk to anyone.
5. The Results: 60% More Site Surveys, Half the Cost Per Appointment
Ninety days in, the numbers moved everywhere that counted. Same lead spend. Same five-person team, just pointed at different work.
Speed to lead: 14 hours to under 60 seconds. Every lead, including the 9 PM ones.
Live contact rate: 76% to 94%. After-hours coverage closed most of the gap.
Site surveys booked: 110 to 176 a month. A 60% increase, with zero extra ad spend.
No-show rate: 30% to 18%. The two-touch reminders did most of that on their own.
Cost per booked site survey: $185 to $72. The AI ran first touch and qualification at $0.35 a minute, so rep hours went to consultations and closing instead of dialing.
Nobody got laid off. The reps just stopped being dialers. Instead of spending the morning leaving voicemails, they opened the day to a calendar of homeowners who'd already confirmed they own the house and have a south-facing roof. That's a different job, and a better one.
6. How AI Qualifies a Solar Lead in Under Two Minutes
This is where solar splits from generic appointment setting. A booked survey with an unqualified homeowner is worse than no booking. It sends a rep driving across town to quote a roof that can't take panels, or a renter who can't sign.
On every call, the AI confirms the five things that decide whether a solar deal is even possible:
Ownership: renters can't install. First filter, every call.
Roof: age, material, and direction. A 25-year-old north-facing roof under heavy shade isn't a candidate, and there's no point sending someone out to confirm it.
Utility bill: average monthly spend sizes the system and the savings pitch.
Credit band: financing eligibility, asked plainly, without the awkward dance a human rep sometimes does around money.
Timeline: buying this quarter or just browsing. That one answer decides routing and how hard the follow-up runs.
It's the same logic we cover in our breakdown of lead pre-screening with AI voice agents, tuned to the specific filters solar installers live and die by.
7. Where AI Calling for Solar Doesn't Work
This isn't magic, and it doesn't fit every solar business. Two cases where we'd tell you to hold off:
Commercial and utility-scale deals: multi-stakeholder buying, custom financing, six-month cycles. Those want a human from the first handshake.
Trust-first local markets: if your whole brand is being the family-owned name everyone in town already knows, an automated first call can clash with that. Test it on a slice of leads before you roll it out.
And the obvious one: AI can't fix bad data. If half your list has wrong numbers, no dialer, human or otherwise, saves you. Clean the inputs first.
8. Key Takeaway: The Bottleneck Was Timing, Not Talent
The installer's reps were good. The problem was that no human team can answer 850 leads a month in under a minute, all night, with a perfectly consistent qualification script. You can't train your way past that. It's a physics problem, and software is good at physics problems.
Same lead budget. 60% more site surveys. Cost per appointment cut by more than half. The leads were always there. They just needed someone to pick up at 9 PM.
Want the B2B version of this story? We documented an agency that replaced three SDRs with AI calling and tripled its pipeline. Different industry, same lesson about timing.
Frequently Asked Questions
What is AI calling for solar and how does solar appointment setting work?
AI calling for solar uses AI voice agents to phone every new lead within seconds of an inquiry, hold a natural conversation, qualify the homeowner on ownership, roof, utility bill, credit and timeline, then book a site survey straight into a rep's calendar. Solar appointment setting is that booking step: the AI confirms the homeowner is a real candidate and locks in a time, so reps walk into a day of pre-qualified visits instead of cold dialing.
How quickly should I respond to a solar lead?
Within five minutes, ideally within one. Leads contacted in the first five minutes are about 21x more likely to convert than those reached at 30 minutes, and a lead loses roughly 80% of its value after the first five minutes go unanswered. The average solar company takes around 47 hours, which is exactly why the fast responder wins the deal.
How can I reduce solar appointment no-shows?
Two-touch automated reminders help most. The installer in this case study sent an SMS and a short voice reminder 24 hours and 2 hours before each survey, and their no-show rate fell from 30% to 18%. Booking the survey live on the first call, while interest is high, beats booking it days later over email.
How many solar leads are unqualified, and can AI filter them?
Roughly 60 to 70% of purchased solar leads are unqualified: renters, heavily shaded or north-facing roofs, or credit below financing thresholds. An AI voice agent screens for all of that on the first call in under two minutes, asking the same questions in the same order every time, so reps only spend time on homeowners who can actually install.
Can AI voice agents qualify solar leads without a human rep?
Yes, for the first touch and qualification. The AI handles dialing, the conversation, and the booking. A human consultant takes over for the site survey and the quote, where on-site judgment and relationship matter. The AI does the volume work people physically can't, around the clock, and hands off only the leads worth a rep's time.
How much does AI calling for solar cost?
TopCalls runs at $0.35 per minute, all-inclusive: telephony, AI processing, and CRM sync are bundled in, so there's no separate Twilio bill or per-seat license. In this case study, cost per booked survey dropped from $185 to $72 once AI took over first touch. For a full walk-through of the math, see our AI cold calling ROI breakdown.
Want to see what AI calling for solar could book for your team? Talk to us and we'll map it to your lead volume and your calendar.
Frequently Asked Questions
Get AI calling tips in your inbox
No spam. One email per week with actionable sales automation tips.
