Cost Per Lead by Channel, Compared Honestly

Cost per lead is what you spent on a marketing channel divided by the number of leads that channel produced. Spend $2,000 on a channel that brings you 25 calls and each of those calls cost you $80. For local service businesses the number usually lands between about $50 and $180 depending on the channel, as of 2026. That range is close to useless on its own, because some of those figures count a phone call and some count a job you actually booked.

Key Takeaways

• A lead costs a local service business roughly $50 to $180 depending on the channel, and the spread inside a single channel is usually wider than the spread between channels.
• A cost per lead and a cost per booked job are different numbers. Most published comparisons mix them, which makes cheap channels look expensive and expensive ones look cheap.
• Referral work and Google Local Services Ads beat owned search on both price and speed, and an owner can run Local Services Ads without hiring anyone.
• Paid channels hold their price forever, because you re-enter the auction every morning. Owned channels get cheaper per lead as volume climbs.

What is cost per lead, and how do you calculate it?

Take everything you spent on one channel in one month and divide it by the leads that channel produced. Spend $3,000 and get 20 calls, and you paid $150 a lead. Spend the same $3,000 and get 60 calls, and you paid $50. Same money, same month, very different business.

The arithmetic is easy. The denominator is where it goes wrong, because nobody agrees on what a lead is. One platform counts a form submission. Another counts a phone call that lasted more than thirty seconds. A third counts a text from someone who wanted a price on a job you do not do. Two shops can report the same number while counting two different events, so your figure and somebody else's average are not comparable unless you know their definition.

Your website matters here more than owners expect: the same traffic at twice the conversion rate costs half as much per lead. Nothing about the channel changed. The page did.

One more distinction the rest of this piece leans on. A lead is an inquiry. An acquisition is a paying customer. A cost per booked job counts work on the calendar. Three denominators, shuffled together constantly.

A second set of eyes on how your numbers are counted is a good reason to book a call.

Why most channel comparisons are rigged

Almost every cost-per-lead benchmark a contractor finds was published by a company that sells one of the channels in the table. The advertising agency's table shows ads winning. The SEO agency's table shows organic winning. The mail house has data proving mail works. We are one of those companies, so treat this page the same way and check the sources at the end of each row.

The second problem: most of the benchmarks that rank for this search are not about you. The marketing software glossaries at the top of the results quote $20 to $50 a lead for software companies and $5 to $15 for online stores. Nothing in those ranges describes a business that replaces water heaters.

Freshness is the third issue, and the one people miss. The most widely cited industry table here, the one giving average lead costs for HVAC and construction, is built on data collected between January 2022 and June 2025 and was last updated in May 2025. It is still worth reading. It is not current-year data, and anyone presenting it as such is careless with your money.

So here is the table with the channels that beat us left in it. Ask us to run it against your own numbers whenever you like.

What a lead costs on each channel

The third column is the one that matters. These figures did not come from one study but from five sources counting five different events, and one row is arithmetic we did ourselves. Take the direct mail line: at roughly $0.50 a piece all in, 5,000 pieces cost about $2,500, and a 1% to 2% response gets 50 to 100 replies. We assumed that 1% to 2%, under the ANA and DMA's 2025 prospect-list rates, and a mail response is not defined the way a phone call is, so read that row as a ceiling, not a forecast.

Two rows carry no number, and that is honest rather than a gap we hid. Referral and AI recommendations have no price per lead because there is no auction to buy into and no vendor with a reason to measure them. Both still belong here, because they are where a healthy service business gets most of its work, and a comparison that drops them flatters the paid channels.

Cost per lead is not cost per booked job

This mistake costs owners real money. Local Services Ads leads average about $53 each in SearchLight Digital's February 2026 data and book at roughly 31%, according to Blue Grid Media's analysis of its own managed accounts plus public benchmarks, verified June 2026. Divide one by the other and the cost per booked job is around $170, not $53. The same source puts Angi at $542 a booked job and Thumbtack at $250, because a shared lead books far less often.

Now put those two numbers in the same column. A $53 sticker price next to a $542 figure looks like a tenfold difference. In booked-job terms the real gap is closer to three to one. Still worth acting on, but you just made a budget decision on a number off by a factor of three.

The pay-per-lead platforms make this easier to check than the pay-per-click ones. Google's documentation on how Local Services Ads bill says you are charged for each valid lead, that leads judged invalid or low quality are not charged, and that charged leads can be credited later if the models decide they were poor quality. Credits are limited to the United States and Canada and are not offered in health care or tax verticals.

Before you compare two channels, ask what event each number counts. If you cannot tell from the page you are reading, keep them out of the same column. How often those leads become booked work also depends on whether anyone picks up the phone, which is its own arithmetic and its own article.

Whenever you want this applied to your accounts instead of to averages, that is what a call is for.

The channels that beat what we sell

Referral and repeat work win. There is no advertising cost, the close rate beats everything else in the table, and you do not need an agency to get it. The honest limits are that it does not scale on demand and it is not actually free, because the cost shows up as your time, your warranty work and the jobs you do at a discount for people who send you business. When the phone goes quiet in February, referrals are not a lever you can pull.

Google Local Services Ads win on speed and price. In every dataset we found they produce cheaper leads than Google Ads search, and the gap widens once you look past the lead. SearchLight Digital's February 2026 analysis of $6.72 million in Local Services Ads spend across 888 contractors and 126,650 leads put the cost per paying customer at $233 for Local Services Ads against $472 for Google Ads. If your phone needs to ring next month, that beats anything organic, including the work we do.

Lead marketplaces have one defensible use. A brand new business with no reviews and no website has nothing else that produces a call this week, and a marketplace does. Treat it as a bridge and build something you own while you stand on it.

GetLocalLeads.AI is an AI visibility and digital marketing agency, and it does not run advertising campaigns. Naming channels we do not sell costs us nothing, which is why you should weigh it more heavily than the rest of this page. To talk through which of these fits your situation, a call is the place to start.

What should a lead cost in your business?

Every benchmark on this page is somebody else's average. How much a lead costs in your business comes out of three numbers you already know: average job value, gross margin and close rate.

Work a roofing example. A $6,000 job at 35% gross margin leaves $2,100 in gross profit. If you close one lead in four, each lead is worth $525 in gross profit before you spend anything on marketing. Decide what share of that you will hand over, and you have a ceiling. At a quarter, the ceiling is about $130 a lead. That comfortably clears the $53 Local Services Ads average and sits well under the $280 First Page Sage reports for a paid construction lead, from data collected between January 2022 and June 2025.

Run the same arithmetic on a $280 drain-cleaning call and the ceiling drops under $20, which rules out almost every paid channel in the table. This is why the same $150 lead is a bargain for one trade and ruinous for another, and why a published average cost per lead is a starting point rather than a verdict.

One lever gets forgotten here. If a first job reliably turns into repeat work and referrals, that first lead is worth more than the single job suggests, and you can afford to pay more for it than a competitor who never calls anybody back. That is a longer conversation about lifetime value.

Setting that ceiling and then holding every channel to it is the job of a fractional CMO (a part-time marketing executive), and the kind of thing our executive consulting work is built around.

Why owned channels get cheaper and paid ones never do

A paid lead is priced by an auction you re-enter every morning. Your cost is set by the second-highest bidder in your market, and nothing you built last year lowers today's price. Stop paying and the leads stop the same day. That is not a criticism of ads. It is how a rented channel works.

Owned visibility inverts the arithmetic. The cost is roughly fixed each month, so as your profile, your site and your content produce more calls, the price of each one falls. The same work that produced 10 calls a month at $300 each produces 40 at $75, and the bill did not change.

The honest cost is time. The industry table cited earlier puts organic at $69 a lead against $115 paid for HVAC, using data collected between January 2022 and June 2025, and the same source is clear that organic takes a longer lead time to get there. Anyone promising you next-week results from local SEO is selling. A lead generation website and a well-fed Google Business Profile are slow to start and cheap to keep.

One more reason this matters now. When someone asks ChatGPT or Google's AI answers for a plumber, they get a shortlist of two or three names, and there is no auction to buy your way onto it. A free AI Visibility Audit is a straightforward way to see whether you are on those lists today.

How do you track this without a spreadsheet project?

You need less than you think. A separate tracking number or source tag for each channel, a "how did you hear about us" question somebody actually asks, and one count a month. That is the whole system. Most owners who think they have no data have two of these three already and have never added them up.

Two rules keep the numbers honest. Measure slow channels on a trailing 90 days, because one month of organic data is noise and you will kill something that was working. And never manage to a blended average, because a blended number is where the one channel burning your money goes to hide. A shop paying $80 a lead overall can be paying $40 on one channel and $220 on another, and the average never says so.

Our clients see this in a live dashboard with a monthly call to go through it, mostly to catch the channel that quietly drifted.

Frequently asked questions

What is a good cost per lead for a home service business?

There is no single good number, and nobody credible will give you one. Across local service channels the figures on this page usually land between about $50 and $180 a lead, depending on the channel and what each number counts. The only benchmark that decides anything is the ceiling you calculate from your own job value, margin and close rate. A call is the fastest route to that number.

How do you calculate cost per lead?

Divide what you spent on one channel over one period by the leads it produced in the same period. The trap is the denominator: decide what counts as a lead first, and apply the same definition to every channel.

Are Local Services Ads cheaper than Google Ads?

In the data we found, yes, on both measures. SearchLight Digital's February 2026 analysis put Local Services Ads at about $53 a lead against $104 blended for Google Ads, and $233 per paying customer against $472. Your market and trade can move that, so verify it in your own account.

What is the difference between cost per lead and cost per acquisition?

One counts inquiries. The other counts customers who paid you. The gap between them is your close rate, so a channel with a low price per lead and a bad close rate can cost more per customer than an expensive one.

Why is my lead cost going up?

Three usual causes: more bidders entered your market's auction, your website is converting a smaller share of the same traffic, or the platform changed what it counts as a lead. Check them in that order.

Where to start this week

Pick the one channel you spent the most on last month. Count its leads with a definition you can defend, divide, and compare that single number to the ceiling your own job value allows. Most owners have never done this for even one channel, and the first time is usually the month a budget decision makes itself. When you want a second opinion on what the numbers say, book a call.