Almost every argument about lead prices is really an argument about someone’s close rate that neither party has stated out loud. A vendor quotes a number, a sales manager says it is too expensive, and nobody writes down the arithmetic that would settle it in five minutes.
The arithmetic is not complicated. What makes it feel hard is that it runs backwards from the sale, and most teams only track the front of the funnel.
Work backwards from the close
The maximum you can pay for a lead is the margin a closed job leaves you, multiplied by the probability that a lead becomes one, multiplied by the share of margin you are prepared to spend acquiring it.
Max lead price = margin per closed job × lead-to-close rate × acquisition share
The third term is a policy decision, not a measurement. Spending every cent of margin on acquisition buys revenue and no profit; spending a conservative fraction leaves room for the funnel to underperform. Pick it deliberately and write it down, because it is the term people fudge when they want to justify a purchase.
A worked example
Take a company with $4,000 of gross margin on a closed job that converts 4% of purchased leads, and a policy of spending at most a quarter of margin on acquisition.
- Value of a lead: $4,000 × 4% = $160
- Ceiling at a 25% acquisition share: $160 × 0.25 = $40
- At $40 a lead, 25 leads produce one job costing $1,000 to acquire against $4,000 of margin
Now change one input. If the close rate is 2% rather than 4% — the difference between exclusive and heavily shared leads is often this large — the ceiling halves to $20. A vendor charging $35 is a bargain in the first case and a slow loss in the second, at the same sticker price.
The figures above are illustrative placeholders chosen to show the method. Substitute your own margin and close rate — the conclusion is frequently the opposite of the one the round numbers suggest.
The rates that actually move the ceiling
Because the terms multiply, the ceiling is far more sensitive to funnel rates than to price negotiation. Three rates dominate:
- Contact rate. A lead you never reach converts at zero regardless of quality. This is where bad contact data destroys lead economics silently — the lead was fine, the phone number was not.
- Sit rate. Appointments that do not hold consume the same acquisition cost and produce nothing. The structural fixes are in solar appointment setting.
- Close rate on sits. The most-coached and least-improvable of the three in the short run.
Haggling a lead from $40 to $35 wins 12%. Lifting contact rate from 25% to 35% wins 40%, and it applies to every lead you buy from now on.
Count the costs the invoice leaves out
The purchase price is rarely the real cost. Fully loaded, a bought lead also carries the setter time to work it, the closer time to sit it, and the drag of duplicates and bad numbers that were billed as deliverable.
This is where shared leads lose most of their apparent advantage. Speed wins them, so working shared leads properly requires immediate response capacity — staffing you are paying for whether or not leads arrive. Put that labour in the model and the cheap channel often stops being cheap.
Bought leads and self-built lists sit at opposite ends of this trade: high per-unit cost and low labour against near-zero per-unit cost and high labour. Compare them on fully loaded cost per closed job or you will systematically flatter whichever one you are already using.
Where Scout Data fits
We sell data rather than leads, which puts us on the self-built side of that trade and we would rather say so plainly. Atlas builds homeowner audiences from live property signals — storm footprints within hours, more than a million new permits a week, roof age, system age, ownership changes — with phone numbers matched by name to the owner of record and scrubbed against the federal do-not-call registry, dead numbers replaced. The per-record cost is a fraction of a lead price; the labour to work it is yours.
Whether that is the better buy depends entirely on whether you have a floor or a crew to work it. A two-person shop with no setter is usually better off buying leads. A staffed call centre is usually better off buying data — and the arithmetic above is how you tell which one you are.
Running it as a habit
Recompute per source, monthly. Lead quality drifts, close rates move with seasonality and staffing, and a channel that cleared the bar in spring may not in autumn. A source that has been below its ceiling for two consecutive months is not a negotiation; it is a cancellation.
Keep the tracking honest by tagging every opportunity with its source at creation, not at close — attribution reconstructed after the fact reliably flatters whichever channel the person doing the reconstruction already believes in.
For the buying side in more depth see how to buy solar leads and aged solar leads. For the equivalent arithmetic on data rather than leads, see the skip tracing cost calculator.
Frequently asked questions
What should a solar lead cost?
There is no correct market price, only a correct price for you. It falls out of three of your own numbers: how much margin a closed job leaves, how often a lead becomes a closed job, and what proportion of that margin you are willing to spend on acquisition. Two companies in the same city can rationally pay very different amounts for identical leads because their close rates differ.
Why do shared leads cost less but often perform worse per dollar?
Because the discount is usually smaller than the drop in close rate. A lead sold to four contractors is not a quarter as valuable — it is worth whatever your odds are of winning a race you entered at the same moment as three competitors, on a homeowner now fielding four calls. Run both through the same arithmetic rather than comparing sticker prices.
Should I buy leads or build my own lists?
Compare them on the same denominator: fully loaded cost per closed job, including the labour to work each. Bought leads carry higher per-unit cost and lower labour; self-built lists carry near-zero per-unit cost and much higher labour. Most teams that run the comparison honestly end up with a mix, and the mix shifts toward self-built as headcount grows.