Your LSA Cost Per Lead Is Lying to You

Every contractor running Local Service Ads watches the same number. Cost per lead. Google puts it right in the dashboard, it is easy to find, and it feels like the score.

It is the wrong number. And watching it is the reason most contractors cannot tell you whether LSA is actually making them money.

Cost per lead does not know if you got paid

Here is all that cost per lead measures: how much Google charged you to make the phone ring. That is it. It says nothing about whether the caller hired you, whether the job was worth your time, or whether you made a dollar.

Two things make it even shakier than it looks.

First, Google records every LSA call and runs a timer on it. The moment a connected call passes 30 seconds, it bills as a lead. So a caller who reaches your team, talks for a minute, and turns out to be price shopping or asking for a service you do not even offer still costs you full price. You were not charged for an interested buyer. You were charged for a timer hitting a number.

Second, and this is the big one. Your cost per lead can be identical to the contractor down the road who closes half as many of those calls as you do. Same number on the dashboard. Completely different reality. Cost per lead cannot see the difference, because it stops counting the moment the phone rings.

“I’ll just get the bad ones credited”

A lot of owners think they can credit their way out of this. Here is the catch. Google’s automated system catches and credits some of the clear-cut stuff on its own, like obvious spam, often in a few days (we see about 72 hours). But a caller who wanted a service you do not offer, or someone outside your service area who called anyway? That one is on you. There is no button to win it back. You cannot credit your way to a healthy margin. You have to operate your way there.

The number that actually matters

The real number is cost per booked job. What did it cost you, in Google fees, to land one actual paying job.

The formula is simple:

Cost per booked job = cost per lead ÷ your close rate

If you pay $50 a call and you book one out of every three callers, each booked job cost you $150 in lead fees. Not $50. You paid for three calls to land one job.

Now you have a number you can use. Put it next to what a job is worth. If a booked job costs you $150 and the job pays $650, LSA is working. If it costs you $150 and the job pays $200, you are running on a treadmill.

Same cost per lead, completely different business

Two electricians, same town, both running LSA, both paying $50 a call.

Ten calls come in for each of them. Same ten.

The first one answers fast. Two of those callers are no-fits, wrong service or outside his area, and he qualifies them out and gets off the phone inside 30 seconds, so Google never charges him for those. On top of that, every lead that comes in gets graded, and two junk calls get flagged and credited back. So out of ten calls, he pays for six, and he books three of them

The second one lets the phone ring. Two callers hang up fast, no charge, same as the first guy. But of the eight that connect, he only answers six. The other two leave voicemails and land right on his bill. He pays for eight, and he books two.

First guy’s cost per booked job: $100. Second guy’s: $200.

Here’s the part that should bother the second guy. He isn’t worse on the phone. When he picks up, he books about the same rate as the first guy. The whole gap is two things he never thinks about: he’s not answering, and he’s paying full freight for calls nobody ever should have. His real cost to land a job is double, and he has no idea, because he’s watching cost per lead, and cost per lead looks fine.

Why almost no contractor can do this math

Cost per booked job equals cost per lead divided by close rate. Google hands you cost per lead for free. You already know what a job is worth. The number you are missing is your close rate, the fraction of callers who actually become jobs.

And you can only know your close rate if someone tracks which calls turned into booked work. Most contractors never do it, so the one number that makes the whole calculation possible is the one nobody has. That is why LSA feels like a gamble. It is not that the channel is random. It is that they are flying blind on the only math that tells them whether it is working.

The 30-Second Siphon: What We Found in 1,700 Real Calls

We audited 1,715 real contractor LSA calls. On the dashboard even though everything looked normal we discovered that 22.5% of them were pure junk that still got billed full price. Wrong numbers, out-of-scope trades, solicitors, people looking for a job. Every one ran past 30 seconds, so Google’s meter charged the card.

Two blind spots caused it.

The answering blind spot. Exactly half, 50.3%, of those billed junk calls lasted between 30 and 60 seconds. A live person picked up, there was a “hi, how’s it going,” a slow “where are you located,” and by the time anyone realized it was a dead end the clock had hit 36 seconds. The charge already landed.

The feedback blind spot. These contractors were not rating their bad leads or archiving the junk, so Google’s system had nothing from them to learn from. Google says your lead feedback helps it send you better-suited leads over time. Starve it of that signal and the quality of what it sends never improves. (To be clear about what this is: rating leads feeds the quality system. It does not let you retarget your own zip codes, and leaving a lead unmarked does not make Google think it was great. Anyone who tells you otherwise is guessing.)

Put the first one in dollars. At 98 billed leads a month at $68 each, getting those short junk calls off the line before the 30-second mark is worth $650 to $1,000 a month.

💰 The Operational Dividend

That is $8,000 to $12,000 a year recovered out of nothing but how the phone gets answered. Better lead quality from feeding Google real feedback comes on top, and builds over time.

And not by being rude. Google pays attention and grades how you handle calls, so a brush-off can cost you ranking and hurt your visibility. You qualify on purpose: open with “Thanks for calling [Company], we do [trade] in [area], what are you looking to get done,” and the wrong-fit caller takes themselves off the line in seconds. The junk drops before the charge. The real buyer stays on, and that higher close rate is what actually drops your cost per booked job on every lead you pay for.

Google Is Watching How You Handle Leads

Responsiveness is one of the signals Google ranks you on, and it is paying attention. Miss calls and let them go unanswered, and Google reads your shop as unresponsive. It will quietly pull back your local exposure (LSA) and route those hot calls to the contractor across town who picks up in two rings.

So flying blind does not just cost you the price of one call. It costs you premium visibility over the next month. The owner who answers, qualifies, and tracks pulls ahead twice: once on the job he books today, and again by building a strong lead history that encourages Google to route more calls his way tomorrow.

This Is an Operations Problem, Not an Ads Problem

Most marketing agencies treat LSA like a traditional ad account. They set the budget, watch the cost per lead, and send a monthly report full of vanity metrics. That is the easy 10 percent.

The 90 percent that decides whether you take home a profit happens after the phone rings. How fast your team answers. Whether the call booked. Whether anyone tracked it. Whether you know your real cost per job by job type, so you can route budget toward the high-margin work and choke off the work that loses money.

That is not advertising. That is running a tight operation. And it is the exact piece traditional agencies leave on your shoulders.

How to Take Control of Your Numbers

You can build this tracking yourself. A whiteboard, a spreadsheet, or your CRM will do it, as long as your team marks every booked job without fail and you run the math every month.

Or you can use a system built to automate it. The Olly Olly LSA Manager grades every call automatically, reads what was said to work out which ones booked, and pulls the date, the service, and the customer name from the transcript. Booked jobs get paired to your spend, so your close rate is an actual number, not a guess, and your true cost per booked job by job type lands on one scoreboard.

Cost per lead tells you what Google took from your card. Cost per booked job tells you if you won the week.

Get a Clean Look at Your Market

If you want to see exactly how your market shapes up, we will run a footprint analysis for your territory. We will show you what the top three contractors in your trade are paying per lead, their estimated call volumes, and the benchmarks you need to hit to beat them.

No sales pitch, no high-pressure close. Just the baseline numbers you need to stop flying blind.