Cost of Marketing

Our Cost Per Lead Quadrupled. The Business Got Healthier.

If a vendor showed you a cost-per-lead chart that climbed from $85 to $371 over five years, you’d fire them before the slide finished loading. That chart is ours. Over the same five years, the close rate hit an all-time high, cost per sale turned downward, and return per marketing dollar posted its best number since 2022.

One of those charts is lying to you about the health of the business. It’s the famous one.

Cost per lead is the easiest number to win and the worst one to chase

Here’s the structural problem with CPL as a scoreboard: every bad decision improves it. Buy shared aggregator leads — CPL drops. Loosen intake standards so anything with a phone number counts — CPL drops. Count branded searches your reputation earned as paid wins — CPL drops. Meanwhile every good decision worsens it: pre-qualify harder, invest in brand channels whose payoff arrives as next year’s cheap demand, walk away from junk volume. A vendor who promises to lower your cost per lead is making the easiest promise in marketing. Most of the ways to keep it are bad for you.

Why ours rose — the honest decomposition

Three forces, and we won’t pretend to know the exact split. First, the chosen one: we deliberately traded lead volume for lead quality — fewer, warmer, more expensive leads that convert at nearly double the old rate. Second, the brand tax: TV, radio, and content channels carried above the per-lead efficiency of pure response channels because their job is making every other channel cheaper later, not winning this month’s CPL contest. Third, the market itself: leads genuinely cost more in 2026 than 2021 across every trade — the published benchmarks say so for everyone, not just us. Any honest reading of a rising CPL has to separate those three before assigning blame, and almost no dashboard does.

The part the comeback version would skip

2021’s ratios were genuinely better — 7.6× return, $734 per sale, the cheapest leads we’d ever see. We didn’t beat that machine; we replaced it, because it was built on demand that was getting exhausted and channels we rented rather than owned. The middle years were expensive: return per dollar slid four straight years to the 5.3× low in 2025 while the new foundation — brand, content, quality systems — was bought at full price. We’ve published that COM% peak honestly elsewhere and we’ll say it again here: the transition was not free, and anyone who tells you theirs was is editing the ledger. What earns the story its ending is 2026: cost per sale falling, return climbing back to 6.9×, and a cost of marketing back inside goal — on the new foundation, not the old one’s leftovers.

Cost per lead prices an input. Cost per sale prices the outcome. When they move in opposite directions, believe the one that cashes.

The two-minute test for your dashboard

Pull your cost per lead and your cost per sale for the last three years and put them side by side. If both rose together, you have a real efficiency problem — go find it by channel. If CPL rose while cost per sale held or fell, your funnel got smarter than your dashboard, and the panic meeting about lead costs is about to make a good system worse. And watch the reverse trap: a falling CPL alongside a rising cost per sale is the most expensive chart in contracting — it means you’re buying more of what doesn’t close. The full grading framework lives here: COM%, the number that governs a contractor marketing budget.

Our cost per lead quadrupled. So did almost everything we’d want a buyer to see. Read the rows that cash.