A solar call center is five functions that each fail differently, run by a manager who is usually only watching one of them. This is the map: what each function owns, the way it breaks, the number that tells you it has broken, and the deeper guide for fixing it. It is written for the person running the floor, not the person selling to it — with one disclosure, below, about where we sit.
Disclosure: Scout Data is our product and it lives in the first function on this page — we sell the homeowner file a floor dials. The other four functions we do not sell anything for, and the advice there is the same we would give a floor that never buys from us.
The five functions
| Function | Owns | Fails as | First number to check |
|---|---|---|---|
| Data supply | Which households get dialled, and whether the number is theirs | Wrong-party conversations that look like weak reps | Right-party rate by source |
| Telephony | Whether the phone rings and what the handset displays | A connect rate that falls on every segment at once | Connect rate by caller ID |
| Compliance | Who may be dialled, when, and the evidence that you checked | A demand letter, months after the dial | Suppression hits per load |
| People | What happens after the pickup | Sets that do not sit | Set rate off right-party, then held rate |
| Measurement | Whether any of the above can be told apart | A blended number that moves and nobody knows why | Whether dispositions are one-outcome-one-code |
The order is deliberate. Each function constrains the one below it — a floor cannot coach its way out of a file matched to the wrong household, and cannot measure anything if the dispositions were accepted as a vendor default — so problems are diagnosed top-down even when they are noticed bottom-up.
1. Data supply
The file decides two of the four rates before a rep says hello: whether the number rings a working line, and whether the person who answers owns the house. Both are properties of how the file was matched, not of the floor, and both are invisible in the match rate a vendor invoices.
Three decisions live here. Whether to buy leads or buy data, which is an arithmetic question about labour — see how to price a bought lead for the model. Which vendor, which is a question you settle on your own records rather than on a sales call — the skip tracing cost calculator converts any two quotes to cost per dialable contact, and how to audit a lead vendor is the afternoon’s work that checks what arrived. And the routine that keeps the file dialable after it lands — list hygiene for call centers — which is where most recoverable performance actually sits.
The failure signature: reps are talking and nobody is booking, and the conversations die on “who am I speaking with.” That is a data problem presenting as a script problem, and coaching will not touch it.
2. Telephony
A solar floor is a small telecom operation with a sales team attached. Carrier relationship, call attestation, a managed pool of numbers with per-number health, and a dialing mode chosen for the floor’s size — those four decisions are made once, badly, during install, and then blamed on the reps for a year.
Outbound dialer setup for solar floors is the build order, from telephony up through dispositions. Which platform to run it on is a procurement problem — CallTools alternatives and Convoso alternatives cover normalising quotes that arrive on different line items, and ReadyMode alternatives works the per-seat arithmetic for a smaller floor.
The failure signature: connect rate fell across every segment, including a brand-new file, in the same week. No list does that. A flagged caller ID does.
3. Compliance
Operational practice, not legal advice. Put your specific dialing setup in front of compliance counsel and treat their answer as authoritative over anything here.
Four lists — the federal registry, state registries, your own internal do-not-call file, and the commercial litigator lists floors at volume choose to buy — plus a calling window that runs on the homeowner’s clock, plus the records that prove the scrub ran. The expensive failures are not exotic: a scrub run before the phones were appended, a national list dialled on the floor’s local time, a March opt-out that a June purchase did not know about.
DNC compliance for outbound solar is the operating routine. Building a suppression list covers the one file that is yours alone and never shrinks, and what is a litigator list explains the commercial product floors add on top. The federal Telemarketing Sales Rule also sets a safe harbor for abandoned calls — no more than three percent of calls answered by a person, measured per campaign over a 30-day period — which is why the dialer’s abandonment ceiling belongs in this function rather than in telephony, even though the pacing knob lives there.
The failure signature: nothing, for months. Then a letter.
4. People
A setter has one job: earn a specific time on the calendar with the people who can say yes. Everything about the role follows from that — the opener leads with the house rather than the product, qualification happens before the booking rather than after, and the handoff carries the context the closer needs to avoid re-opening the conversation from zero.
Solar appointment setting is the operational version of the role. Solar cold call script is the conversation in parts, and solar rebuttals and objection handling is what to say when the opener lands and the homeowner pushes back anyway. How setters are paid shapes what they book — commission structures for solar reps covers why a plan that pays on sets buys sets, and a plan that pays on sits buys sits.
The failure signature: sets are up and the closers are furious. A floor that measures sets will always produce sets.
5. Measurement
The only function whose failure makes the other four undiagnosable. Its whole substance is three rules: one disposition per outcome and no code two agents would use differently; every record stamped with its source and batch at import, so nothing is ever blended; and one written definition per metric, dated, never silently changed.
With those in place, the floor reports a chain rather than a total — dials, connects, right-party, sets, held — and a bad week has a location. How many dials per appointment turns the chain into the staffing and cost arithmetic; KPI benchmarks for solar call centers covers which report numbers flatter and which to watch instead; what is a good contact rate explains why the first link in the chain has four definitions and why you must pick one. And per source, monthly, how to measure lead source ROI is the ledger that decides which files get reordered.
The failure signature: a number moved, three people have three explanations, and the meeting ends with a coaching plan.
Where the money goes
Every cost on a floor is a cost per held appointment once you divide it through, and that is the only unit on which data, seats, telephony and people can be compared. The shape of the trade:
- Data is priced per record and paid up front. Cheap per unit, expensive in labour, and the term that sets the top two rates of the chain.
- Bought leads are priced per lead and carry the labour inside the price. Expensive per unit, cheap in labour, and the right buy for a floor that does not yet have setters.
- Seats and telephony are priced monthly, per agent, with carrier fees that some platforms publish and some bill separately — the pricing pages linked above exist because the quotes arrive incomparable.
- People are the largest line, and the one whose productivity is entirely determined by the three lines above it.
The buy-leads-or-buy-data decision is the pivot. A floor with idle setter hours is paying for labour it is not using, and data is the cheaper way to fill them; a floor with no setters is paying for labour inside the lead price and should keep doing so until it has enough volume to justify a seat. Exclusive vs shared leads runs the arithmetic on the lead side, and aged solar leads covers the cheapest tier of it.
A working week
| When | Function | What runs |
|---|---|---|
| Every load | Data, compliance | Standardise, dedupe, suppress, then append, then scrub, then load — in that order |
| Daily | Compliance, telephony | Opt-outs into suppression the same day; abandonment and per-number connect checked |
| Weekly | Measurement, data | The four rates per source and segment age; retire exhausted segments; recompute dials per held |
| Monthly | Data, people | Source ROI ledger; re-scrub anything held; audit a sample for wrong-party; review setter pay against held |
| Quarterly | Telephony, measurement | Hour-of-day report by local hour and line type; re-plan the shift pattern against it |
The hour-of-day pass is the one floors skip, and it is the one that usually shows the best permitted hours were the least staffed — best times to call homeowners has the procedure.
The decisions that are hard to reverse
Most floor decisions are cheap to change. Four are not, and they are worth slowing down for:
- The disposition set. Every report you will ever run is built on it, and history recorded under a bad one cannot be repaired.
- Where the suppression file lives. Inside a dialer you might replace is inside a dialer you will lose it with.
- Whether sources are stamped at import. A blended file can never be un-blended, and a vendor you cannot evaluate is a vendor you cannot fire.
- What setters are paid on. Sets or held. The floor will produce whichever one you chose, at volume, for as long as the plan stands.
Get those four right in the first month and the remaining decisions are experiments with a stop condition. Get them wrong and the floor will spend a quarter unable to tell which function it is arguing with. For what we ship into the first function, see what we deliver to call centers.
Frequently asked questions
What does a solar call center actually do?
It converts a file of homeowners into held appointments for closers, at a cost per held appointment the business can afford. Everything else — the dialer, the scripts, the compliance stack — exists to move one of the four rates between a dial and a held appointment, and a floor manager’s job is to know which of the four is the problem this week.
Should a solar call center buy leads or buy data?
A staffed floor is usually better off buying data, because it already has the one thing a bought lead is priced to replace: the labour to work a record. A two-person shop with no setters is usually better off buying leads. The arithmetic that decides it for your floor is cost per held appointment, fully loaded, per source.
How many setters do I need?
Work backwards from the held appointments the closers can run in a week. Divide by held appointments per setter-hour — which comes from your own four rates and your dials per hour — and you have setter-hours per week. Do not staff from a published dials-per-appointment figure; there is no honest one.
What is the most common reason a new solar floor fails?
It cannot tell which layer is broken. Connects fall, the manager coaches harder, and the actual cause is a flagged caller ID, an exhausted file, or a suppression step nobody wired in. Floors built so that each function is measurable on its own get a diagnosis; floors built script-first get a mood.