Before you spend a dollar on ads, run this math first

Most contractors who got burned by an agency skipped one calculation before signing. Not because they're bad at math. Because the agency never asked them to do it, and they didn't know they should. The number is simple: how many leads do you need each month, at what cost per lead, for this to be worth anything at all?

If you can't answer that before spending a dollar, you're not running a test. You're guessing. And when the results disappoint, you won't know whether the problem was the targeting, the creative, the follow-up, or the math you never did.

The detailer who figured it out the hard way

Samuel runs a car detailing operation he started in year 11. Monthly maintenance packages at $100 per client. He tried Meta ads through another provider on a $20-per-day trial. Leads came in. The trial looked good. When the trial ended, the agency quoted $1,000 per month to continue.

He stopped immediately.

But there was a second problem he almost didn't mention: most of the leads were an hour and a half away. He couldn't service them himself, so he was paying $20 per job to subcontractors and passing the work off. The agency had run broad targeting to make the trial look impressive. Samuel was technically getting leads. He was losing money on most of them.

This is the failure mode almost nobody talks about. The lead count is real. The cost-per-lead looks fine on paper. But the serviceable lead count, after you filter for geography and factor in what it costs to fulfill jobs you can't reach yourself, is a completely different number.

How the math actually works for a recurring-revenue model

Samuel's business runs on monthly maintenance clients. That changes the calculation in a way most lead-gen conversations ignore entirely.

A one-time detail might be worth $150. A monthly maintenance client at $100 per month, retained for a year, is worth $1,200. Those are not the same lead. An agency optimizing for raw lead volume will treat them identically. You shouldn't.

If a monthly maintenance client is worth $1,200 over twelve months, you can afford to pay more per lead to get one. The floor on acceptable cost-per-lead is higher. The tolerance for a longer conversion timeline is higher. But here's where it gets specific: if your close rate on leads is 20 percent, you need five leads to get one client. At $1,200 lifetime value, you can spend up to roughly $240 acquiring that client and still be in the black, before ad spend eats your margin.

That means your maximum viable cost-per-lead is somewhere around $40 to $50, assuming a 20 percent close rate and a one-year retention. At $10 CPL you're profitable even if your close rate is half that. At $100 CPL with a 20 percent close rate, you're spending $500 to acquire a $1,200 client. Tight. One month of churn and you've lost money.

None of this math is exotic. But it has to be done before you start, not after you've already paid.

What changes when you subcontract out of range

Here's what Samuel's numbers looked like in practice. The agency gave him leads. He was getting one to four per day during the trial. Some he could service himself. Many he couldn't. For those, he paid $20 per job to someone else and passed the work on.

At $20 subcontracted fulfillment cost per job, a $150 detail becomes a $130 job. That's fine on its own. But add a $40 cost-per-lead, a 20 percent close rate, and you've spent $200 in acquisition to make $130 on the job. The math inverts.

This is why geography isn't a nice-to-have in targeting. It's a variable that directly changes whether the unit economics work. A lead 90 minutes away isn't worth the same as a lead 15 minutes away. If your cost-per-lead is based on blended volume across both, you're making a profitable number look real when it isn't.

The right way to structure this before you start: define your serviceable radius, build targeting around it, and calculate cost-per-lead against serviceable leads only. A campaign producing 50 leads per month across a 90-mile radius, where you can realistically reach 20 of them, is a 20-lead campaign. Price it accordingly.

The number to have before you talk to any agency

Before any sales call with any agency, run this calculation yourself:

Your average job or contract value. Multiply by your realistic close rate (be honest, not optimistic). That gives you maximum acquisition cost per client. Divide that by your expected average close rate expressed as a decimal. That's your ceiling cost-per-lead.

For Samuel: $100/month maintenance client, 12-month expected retention, $1,200 lifetime value. Realistic close rate on a warm lead: 25 percent. Maximum acquisition cost: $300. Maximum cost-per-lead: $75. But he also subcontracts some jobs at $20 cost per job, which cuts his effective margin. So the real ceiling, accounting for fulfillment cost on subcontracted work, is lower.

The Safe Step case study from ASN's own results is a useful comparison point: 247 leads, $11 CPL, $2,800 total ad spend. Rubber resurfacing, not detailing, but the math structure is identical. At $11 per lead with a 25 percent close rate, you're spending $44 to acquire a client. For most recurring-revenue home service businesses, that's well inside the profitable range.

If an agency can't show you a cost-per-lead from a comparable trade, and can't explain how they'll constrain targeting to your actual serviceable area, those are the two questions to push on before anything else. Not the guarantee language. Not the contract terms. The CPL from a business with a similar ticket size, and confirmation that the targeting radius maps to somewhere your crew can actually work.

What to do with this math

Take twenty minutes before your next agency conversation and run the numbers above. Lifetime client value, close rate, maximum acquisition cost, maximum CPL. Then add one more line: what percentage of your leads will require subcontracting or be outside your serviceable radius, and what does that cost you per job?

If an agency can match or beat your CPL ceiling with niche-adjacent proof, and they're not asking for a setup fee or a long-term contract before they've shown you anything, the test is low-risk by definition. If they can't show you a comparable CPL and want commitment upfront anyway, the math you just ran tells you exactly what you're risking.

If you want to see how ASN structures campaigns for recurring-revenue service businesses, the contact page is the right next step.