Every lead marketplace quotes you a price per lead. That number is close to meaningless on its own, because you do not win every lead you buy. The number that decides whether the channel works is your cost per job actually won, and it is almost always several times the sticker.
The arithmetic is simple enough to do on the back of an invoice, and worth doing before your next renewal.
The formula
Cost per won job = (price per lead ÷ your close rate on that channel).
If leads cost you $40 and you close one in four, each won job cost $160 in acquisition before you turned a wrench. Close one in eight — which is common when a lead is sold to several contractors and you are the third to call back — and it is $320.
Now take that against your average ticket. On a $250 service call, a $320 acquisition cost means the job lost money before fuel. On a $2,000 install it is a rounding error. This is why the same lead price is a bargain for one trade and a slow bleed for another, and why contractors in low-ticket trades are the ones who churn off these platforms first.
The costs that never show up on the invoice
- Time spent calling leads that never answer, which is unpaid and invisible
- Time quoting jobs against three other companies who each also paid for the same lead
- Fuel and windshield time on estimates that were never going to close
- The credits process — the hours spent disputing bad leads, which is itself a cost even when the credit lands
- The opportunity cost of the actual job you could not take because you were chasing four that did not exist
Why the model produces this outcome on purpose
None of this is anybody behaving badly. It is what the model is. A business that earns revenue when a lead is sold has to sell leads; a business that earns revenue when a job completes has to make jobs complete. Both are rational. They just point in different directions, and the direction shows up in your bank account rather than in the marketing.
The test is straightforward: when a lead goes nowhere, who absorbed the loss? If the answer is only you, the incentives are not aligned with yours and no amount of account-manager goodwill changes that.
What the other structure looks like
Roger charges a percentage of a job that closes, 10% to 30% depending on the trade and the size of the job, and you see the exact fee on the offer before you accept. A job that does not close costs nothing, because there was nothing to take a percentage of. There is no charge for accepting a job, no charge for replying to a homeowner, and no annual contract. If you would rather have a fixed number, there is a flat monthly fee instead — the current bands are on the contractor pricing page.
That is not a claim that Roger sends more work. Roger is early and will not pretend otherwise. It is a claim about who carries the risk of a job that does not happen.