Lead Scoring for Solar Installers: Demographics Beat Gut Feel

Residential solar sales is a brutal business pretending to be a green one.

The average installed system costs $25,000–$45,000 before incentives. The sales cycle is 60–120 days. The cost to put a closer on a kitchen table — drive time, prep, the actual consult, the follow-up — runs $300 to $700 once you load it honestly. Customer acquisition costs in residential solar regularly hit $4,000–$7,000 per closed deal in competitive markets. Margins are thin enough that two bad weeks of misallocated consults can take down a small installer’s quarter.

And almost every installer in America scores leads the same way: gut feel.

The lead came from Google. The voice sounded interested. The address is in a “good neighborhood.” Send a closer. Cross your fingers.

This is not a sales strategy. This is gambling with very expensive chips.

The companies that are quietly running away with the residential solar market in 2025 and 2026 aren’t outclosing anybody on the kitchen table. They’re outqualifying everybody before the kitchen table. They’re using demographic data, property data, and AI voice intake to score every inbound lead in real time, so that the consults that do happen close at 2–3x the rate of their gut-feel competitors.

This is how it works, why it works, and what the math looks like when you stop sending closers to leads that were never going to buy.

The economics nobody wants to talk about

Let’s look at the unit economics of a typical mid-size residential solar installer.

Cost per marketing lead (CPL): $80–$220 depending on channel
Show rate to consult: 55–70%
Close rate at consult: 18–28% (high performers hit 35%)
Net cost of a held consult (including no-shows): $400–$700
Cost per closed deal (blended): $4,200–$7,800

Here’s the part that hurts: across the industry, an estimated 40–55% of consults are run on households that could not have closed under any circumstances. They don’t own the home. They don’t have credit. They don’t have a roof in shape to host a system. They have a planned move inside 24 months. They’re in a state, utility, or net-metering regime that breaks the savings math. They’re shopping six installers and signing with the cheapest one — which is never going to be the careful, full-service installer running the consult.

Every one of those consults cost $400–$700 to run. None of them was ever going to produce revenue. The installer didn’t lose them at the kitchen table. They lost them by sending a closer in the first place.

Now flip the math. If demographic and property-based lead scoring lets you screen out even half of those unqualified consults before you dispatch a closer, your effective cost per closed deal drops from $6,000 to $3,200 overnight. Same marketing spend. Same closers. Same close rate at the table. Half the wasted consults.

That is the entire game in residential solar right now. The companies that figure this out scale. The ones that don’t get bought or shut down.

What “demographic lead scoring” actually means in solar

Most solar installers think they already do lead qualification because their setters ask three questions: do you own the home, what’s your electric bill, when are you looking to move forward.

That isn’t qualification. That is asking the lead to qualify themselves, while motivated to lie.

Real demographic lead scoring uses third-party data layered against the caller’s verified information to score the lead on the dimensions that actually predict close rate. The strongest predictors in residential solar, in rough order of importance:

Homeownership — verified, not stated. Renters cannot install. ~20–30% of inbound “solar” leads in some markets are renters.
Property tenure — how long they’ve owned. Owners with <2 years tenure are statistically less likely to install (they’re still figuring out the house). Owners with 4–15 years of tenure close at the highest rates.
Estimated home value — predicts ability to finance and likelihood of solar being a sensible improvement.
Household income tier — gates financing options and predicts close rate.
Roof age and roof characteristics — older roofs require replacement before installation, which kills many deals or shifts them into a re-roof + solar bundle.
Property characteristics — orientation, size, shading proxies from public data.
Local utility and net-metering regime — automatic ZIP-level scoring. Some ZIPs have terrible payback math; no point burning a consult there.
Age and household composition — predicts decision-making style and timeline (younger dual-income households move faster; older households want more in-person reassurance).
Education and occupation — proxies for how the consult should be framed (technical vs financial vs lifestyle).
Prior solar inquiries (when observable) — repeat shoppers are usually price-sensitive comparison buyers, not loyal closers.

Each of these is a data point a solar installer historically had to guess at, or worse, wait until the consult to discover. With modern caller intelligence, all of them can be present before the AI voice agent finishes saying hello.

How an AI voice agent qualifies on the inbound call

Here is a realistic 4-minute call to a solar installer using Caller Technologies’ AI voice agent.

0:00 — Inbound call routed to AI. Caller intelligence loads in under 800ms. The system identifies: 9-year homeowner, single-family detached home, estimated property value $565,000, household income tier consistent with prime financing, two adults age 41 and 39, roof age estimated at 11 years from county permit records, ZIP code in a state with strong net-metering. Pre-load lead score: high.

0:05 — AI greets the caller by company name, friendly tone, normal pace for the demographic.

0:20 — Caller explains they’re “looking into solar.” AI confirms the address on file (or asks if it’s for a different property — important catch). Caller confirms.

0:50 — AI asks two qualifying questions framed as helpful context: monthly electric bill range (gates payback math) and any major roof work in the last few years (catches the re-roof issue).

1:30 — Caller mentions a $230/month average bill and a roof they “think is original to the house.” AI cross-references: roof age >10 years on a 2014 build, monthly bill in the strong payback range. Lead score updates: still high, but with re-roof flag.

2:10 — AI explains the typical process, asks about timeline. Caller says “next few months.” AI doesn’t push. It offers two consult windows within the next eight days, both pre-routed to a specific senior consultant whose calendar matches.

3:00 — AI confirms the consult, sends an SMS confirmation with a pre-consult video about the company, and triggers an automated email with a roof-age FAQ targeted to this specific objection.

3:45 — Call ends. The CRM record is fully populated, the consult is on the closer’s calendar with a high-confidence lead score, and the marketing system has flagged this lead for premium nurturing if the consult doesn’t close on the first visit.

That’s not a script. That’s a conversation. The caller doesn’t experience qualification — they experience helpful, well-informed service. The installer experiences a lead that’s already 70% pre-sold by the time the closer rings the doorbell.

What scoring looks like for the calls that don’t qualify

The other half of the math is the calls you stop running consults on.

Renter call: AI politely confirms the home isn’t owned, explains solar requires homeownership, offers to keep them in the system for if/when they purchase, and ends the call with no consult booked. Zero closer time burned. Lead is preserved in the CRM in case their status changes.

Owner with planned move: AI hears “we’re listing in the spring.” Doesn’t push for a consult. Offers to send an information packet and schedule a callback in six months when they’re settled in the next home. Consult avoided. Relationship preserved.

Owner in a bad payback ZIP: AI doesn’t tell them they’re in a bad ZIP (that’s a marketing problem, not a customer problem). Instead, it captures the lead, routes to a low-priority nurture sequence, and avoids dispatching a closer who would walk into a difficult savings story.

Tire-kicker with a 24-month timeline: Scored as a long-cycle nurture lead. No consult. Automated drip campaign engaged. If they re-engage in 18 months with stronger intent, the system upgrades them and books the consult then.

Each of these calls cost the installer functionally nothing — no closer dispatched, no driving, no kitchen-table no-show. And each of them is preserved as a future opportunity that gut-feel qualification would have either burned at consult or lost entirely.

The conversation adaptation that closes more

Solar is unusual in home services because the caller’s cognitive profile often matters more than the property profile.

A 34-year-old dual-income software couple wants the financial model in their inbox before the consult, ROI numbers on the table, and a 45-minute consult that respects their time. If you send a closer who runs a 90-minute trust-building presentation, you lose them.

A 68-year-old retiree on a fixed income wants reassurance, slow explanation, multiple visits, and a closer who’ll come back twice if needed. If you send a closer who hands them a tablet and walks them through a fast digital presentation, you lose them too.

An AI voice agent adapts to both within the qualifying call itself — pace, vocabulary, framing — and tags the closer assignment accordingly. The right closer walks into the right house with the right preparation. Close rates climb without any change in marketing, pricing, or product.

The data set behind the scoring

Caller Technologies’ demographic intelligence draws on 2+ trillion data points across 3+ billion people, surfacing up to 150 demographic data points per call. For a solar installer, the meaningful subset includes:

Verified address and property ownership
Estimated property value and equity proxies
Property tenure and prior transaction history
Household composition and ages
Income tier and occupation
Education indicators
Lifestyle and consumption signals
Distance from caller’s address to nearest existing job or service area
Property characteristics (size, type, year built)
Social and contact verification

None of this requires the caller to provide it. None of this slows the call down. All of it is available before the AI even finishes the greeting, and it stays available for the closer who eventually walks the kitchen table.

Real-world results

Solar installers running this stack consistently report:

30–45% reduction in dispatched consults (the unqualified ones are filtered out)
20–35% increase in close rate at the kitchen table (because the consults that are run are pre-qualified)
40–55% reduction in cost per closed deal (the combination of fewer wasted consults and higher close rate at the ones that run)
2–3x increase in setter productivity (one AI handles call volume that previously required a small setter team)
Faster time from inbound to consult (no setter callback delay; AI books in the conversation)

These are mid-line outcomes, not best-case. The math is brutal in solar, and the leverage from getting qualification right is correspondingly brutal in the other direction.

Objection handling

“My setters know how to qualify. They’ve done this for years.” Some of them do. Most of them ask the wrong questions and trust the answers. Caller intelligence verifies homeownership and property value without asking. It scores tenure without asking. It catches the roof age problem before the closer drives an hour to find out. Your setters become 3x more productive because they only work the leads that actually merit human touch.

“I don’t want my brand answered by a machine.” The AI voice agent uses your brand, your voice profile, your scripts, and your tone. Customers consistently report not realizing they’re on an AI call. CSAT is equal to or above human-only intake. The brand experience improves, not degrades.

“My margins are too thin for new tools.” Your margins are thin because of the wasted consults. The installer with healthy margins isn’t outclosing you — they’re outqualifying you. The cost of the platform is a small fraction of the cost of a single unnecessary consult per week.

“What about leads that need a delicate human touch from minute one?” Caller intelligence flags those instantly. High-value referral leads, repeat customers, complex commercial properties, and known partner contacts can be routed directly to specific humans before the AI even greets them. The AI handles the volume floor. Humans handle the relationships only humans can.

The takeaway

Residential solar is going to consolidate over the next 36 months. National players, capital-backed regionals, and a small number of disciplined independents are going to take share from everyone else. The dividing line between the winners and the rolled-up isn’t going to be product, price, or marketing spend. Those are commodities.

The dividing line is going to be what you know about a lead before you spend money on it. The installers who score leads with real demographic data and route them with AI are going to operate at half the customer acquisition cost of installers who still run the gut-feel playbook. Half the CAC compounds. Six quarters in, the gap is unbridgeable.

The good news: the technology to fix this is already in production, already affordable, and already deployed by some of the smartest installers in your market. The bad news: that means it’s already deployed by some of the smartest installers in your market.

Learn how demographic AI works for solar. Caller Technologies will walk through your last 90 days of inbound leads, score them against the same data its AI voice agent uses live, and show you exactly which consults you ran that were never going to close — and which leads you missed that would have. Bring the report to your next sales meeting. The conversation tends to change quickly after that.

See the numbers for your own business with the ROI calculator, or compare plans on pricing.

See who’s calling before you say hello. The Caller Technologies AI voice agent answers 24/7, qualifies every caller with 150+ demographic signals — owner or renter, home value, income — and books real jobs while your crew works. Start your free trial — free until you book a paying job, no credit card.