What should cost of marketing be for a contractor, and why does it keep swinging?

What should a contractor’s cost of marketing be, and why does it move so much?

We run to 10% to 15% of net revenue. Ours ran 14.2% in 2022 and 19.4% in 2025 at the eight-figure Toledo exteriors company where I direct marketing, and the percentage hides that cost per demo went up 2.5 times while the value of a demo went up 1.9 times. Marketing controls the first number. Ticket, close rate and cancellations control the second, and they belong to sales.

What is cost of marketing, exactly?

Everything spent on marketing in a year, divided by the net revenue the company booked in that year, and the ratio is useful precisely because it survives a change in company size. A $5M shop and a $15M shop can both run at 14%. The trouble is that a ratio moves when either half moves, and for about three years I answered for the whole ratio in every leadership meeting while only half of it was mine.

Here is what the halves did at our company, from the Cost of Marketing report in Lead Perfection, 2022 to 2025. Spend went from $1.30M to $2.48M. Net revenue went from $9.1M to $12.8M on that report’s attribution. So the percentage went 14.2%, 16.6%, 17.5%, 19.4%. Read on its own that looks like marketing getting steadily worse at its job, and that reading is roughly half right.

Why did ours go from 14% to 19%?

Cost per demo, the price of getting a rep in front of a homeowner, went from $398 in 2022 to $1,012 in 2025. That is 2.5 times, and it is the marketing half of the story stated plainly. I have written elsewhere about why we let that happen on purpose, because we traded cheap low-intent volume for fewer appointments that closed, and the set-to-sale rate went from 16% to 25% while we did it (A Demo Costs $992 Now).

The other half moved too. Net revenue per demo went from $2,793 to $5,209, which is 1.9 times, and that came from the average net ticket rising every year ($6,040 to $9,457), the close rate climbing from 48% to 55%, and the share of signed jobs that actually installed going from 62% to 83%. None of those is a media decision, and every one of them moves the percentage.

So the demo cost 2.5 times as much and was worth 1.9 times as much, and the gap between those two numbers is the whole 14% to 19% move. That is the arithmetic the ratio hides.

Which number does marketing actually control?

Cost per demo. It has one thing in the numerator marketing chooses, spend, and one thing in the denominator marketing mostly produces, appointments that sat. When I stopped defending the ratio every month and started reporting cost per demo next to net revenue per demo, the conversation with leadership changed from “why is the percentage up” to “the demo got more expensive and it also got more valuable, here is which one moved this quarter.”

I will be blunt about the cost of learning that late: for a couple of years I was running around like a chicken with my head cut off answering for a percentage that moved when pricing changed, when the cancellation rate changed, when a storm year rolled off, and I should have been looking at the two underlying numbers the whole time.

Where this is wrong

Cost per demo alone will lie when demo quality shifts, because a shop that buys cheaper, worse appointments will see cost per demo fall while close rate falls faster, and the percentage will get worse with a cost-per-demo chart that looks like a win. It has to sit next to net revenue per demo or it is half a metric. And the 10% to 15% band is ours, an operating standard for an exteriors company with a canvass operation and a broadcast budget, not a survey result, so a service trade with $1,800 tickets should not inherit it without checking what its own margin allows.

What to do

• Tonight: pull spend, demos and net revenue by year from the CRM and compute cost per demo and net revenue per demo side by side. Whichever one moved is the one to talk about.
• This quarter: put both numbers on the monthly report above the percentage, and put the percentage on net revenue with the pull date on it.
• Ongoing: when the percentage moves, decompose it before reacting. On cost per demo alone ours would have gone to 36%, and the value of the demo pulled it back to 19%.

Sources: Lead Perfection “Cost of Marketing by Source” and “Set Appointment Statistics by Sub-Source,” Toledo, 2022 to 2025 full year, first party. The 10% to 15% band is a Capstone operating standard.

Related notes