A good cost per lead is the most your business can pay for an enquiry and still make a profit, not a number copied from a benchmark table. In US data, the median cost per lead is $66.69 on Google search ads and $27.39 on Meta lead ads, but it runs from about $30 for auto repair to about $132 for legal services. Where you land depends on what a customer is worth and how many leads actually buy.
This guide gives you a formula for your own limit, a worked example, 2026 US benchmarks for Google and Meta, and the limits of those benchmarks. See also our guides to Google Ads cost and Meta ads cost.
What cost per lead means
Cost per lead (CPL) is your ad spend divided by the number of leads it produced. Spend $2,000 and get 40 enquiries, and your CPL is $50.
A "lead" can be a form submission, a call, a message or a booking request, and platforms count them differently. CPL is also not the cost of winning a customer. That is cost per acquisition, and it is always higher, because not every lead buys.
How to work out your own limit
A benchmark tells you what others paid, not what you can afford. Four numbers do.
- Customer valueRevenue from a typical new customer over a period you choose, such as the first year. Be conservative.
- Gross marginThe share of that revenue left after the direct cost of delivering the service. Multiply to get gross profit per customer.
- Share you will spend on adsSpending 100% breaks even on the first sale. Most businesses choose a lower share so ads also leave a profit.
- Lead-to-customer rateThe percentage of leads that become paying customers. Measure it from your own enquiries. Do not guess.
Maximum cost per lead = customer value × gross margin × share spent on ads × lead-to-customer rate
If your actual CPL is below that number, the channel pays for itself on your terms. If it is above, you need cheaper leads, a better close rate or a higher-value customer.
A worked example
These figures are placeholders chosen to keep the arithmetic easy, not a forecast. A home-services business earns $1,200 from a typical new customer, with a 50% gross margin, and is willing to spend half of that margin on ads. One in five leads (20%) becomes a customer.
Gross profit per customer is $600. Half of it, $300, is the most it will pay for a customer. At a 20% close rate, the maximum cost per lead is $60 (break-even, spending all the margin: $120).
| Scenario | CPL | Buy rate | Per customer | Share of profit |
|---|---|---|---|---|
| Google search benchmark | $90.92 | 20% | $455 | 76% |
| Meta lead ads benchmark | $42.95 | 20% | $215 | 36% |
| Meta, only 10% buy | $42.95 | 10% | $430 | 72% |
| Meta, only 5% buy | $42.95 | 5% | $859 | 143% |
Limits: illustrative. CPL figures are the home-improvement medians below; close rates are assumptions.
Google's benchmark is above the $60 limit, yet still profitable: it uses 76% of gross profit, not the 50% planned. Meta's is inside the limit, but only if its leads buy at the same rate. At 10% it uses 72% of gross profit, and at 5% the business loses money on every customer. The CPL looked cheap in all three Meta rows. Only the close rate told the truth.
Maximum cost per lead calculator
Enter your own numbers to get your limit and compare it with a US benchmark.
Maximum affordable cost per lead
Google search benchmark: $90.92 per lead, 52% above your limit.
Meta lead ads benchmark: $42.95 per lead, 28% below your limit.
The starting values are placeholders, not benchmarks. Benchmarks are US dollars: if you choose pounds or euros, enter today's exchange rate so the comparison is fair. Leaving it at 1 treats the currencies as equal.
Benchmarks by industry (US)
These are the latest figures we could verify, from LocaliQ and WordStream, covering April 2025 onwards. Google figures are for search campaigns; Meta figures are for lead campaigns. Some Meta cells are empty because the source has no lead benchmark for that industry.
| Industry | Google search CPL | Meta lead ads CPL | Lower |
|---|---|---|---|
| Attorneys & legal services | $131.63 | — | — |
| Furniture | $106.70 | $39.25 | Meta |
| Real estate | $102.51 | $13.74 | Meta |
| Business services | $93.69 | — | — |
| Home & home improvement | $90.92 | $42.95 | Meta |
| Education & instruction | $77.48 | $26.31 | Meta |
| Industrial & commercial | $75.19 | $35.87 | Meta |
| Finance & insurance | $74.44 | — | — |
| Dentists & dental services | $72.97 | $61.56 | Meta |
| Health & fitness | $67.36 | $27.11 | Meta |
| Personal services | $54.60 | $38.09 | Meta |
| Physicians & surgeons | $40.04 | $32.14 | Meta |
| Beauty & personal care | $39.25 | $50.91 | |
| Restaurants & food | $30.57 | — | — |
| Auto repair & service | $29.96 | — | — |
| All industries (median) | $66.69 | $27.39 | — |
Limits: US only, medians, 2025-26 data. A dash means no Meta lead benchmark in the source. The all-industries rows cover different sets of industries (23 for Google, 14 for Meta), so do not read them as a like-for-like gap.
High-value services such as legal, real estate and home improvement pay the most per lead on Google. Urgent local needs such as auto repair and restaurants pay the least. Beauty is the one industry where Google was cheaper than Meta.
Year on year, PPC Land reports the Google median fell from $70.11 to $66.69 and the Meta lead median from $27.66 to $27.39. Do not count on that continuing: see why paid ads keep getting more expensive.
How to read a benchmark
They are medians, not your average
Both sources call them "averages" but state they are medians, chosen to limit outliers. Half of campaigns cost less, half more.
The sample is US-only and uneven
The Google data covers 13,474 US campaigns with at least 52 per category. The Meta lead data covers 452 campaigns with as few as four per category, so some industry figures may reflect a handful of accounts. LocaliQ says the data comes from its own customers, not every advertiser.
A lead is not defined the same way twice
A Meta form submission and a Google phone call are both "leads", but not the same level of intent.
The numbers do not always reconcile
PPC Land notes the Meta median click cost and conversion rate do not multiply back to the median cost per lead, and the LocaliQ Facebook page contradicts its own table in one passage. Treat precise decimals with caution.
Averages hide your market
City, season, offer, creative and competition all move CPL. Use benchmarks to spot a number that looks wildly off, then switch to your own data.
The UK and other channels
We could not find a reliable published cost-per-lead benchmark for the UK. The most UK-relevant source we opened reports cost per click and cost per conversion in pounds, but it does not state its geographic scope, its pound figures appear to be converted from dollars, and some industries rest on three to eight accounts. We did not use its numbers. In the UK, treat the US figures as a rough guide and work from your own limit.
For LinkedIn, the only data we found was one B2B agency's report, which gives cost-per-lead ranges of $200-$250 for North America and $120-$150 for Europe. It does not explain its method and covers B2B campaigns, not local services, so treat it as a reminder that B2B leads cost far more, not as a benchmark. For TikTok, see our TikTok cost guide.
Lead cost versus lead quality
A lead is only useful if someone answers, books and turns up. The chain looks like this: lead, contacted, booked, attended, paid. Each step loses people, and CPL ignores all of them.
That is why cost per booked customer is the better measure. In our experience, search leads tend to be warmer because the person was looking for the service, while social leads need faster follow-up. That is our observation, not a published finding, so test it on your own account.
You can improve the chain without more ad spend: an instant reply with a booking link, reminders and reactivating old enquiries. Our guides on AI chatbots for booking appointments and reducing no-shows cover both. In our own account of a UK clinic client, Manchester Aesthetics, reactivating existing contacts saved about 15 appointments a week.
How to lower your cost per lead
Fix the offer before the bid
A specific service and a clear reason to act now usually lift conversion more than tweaking budgets.
Send traffic to a page built for one action
One service, one promise, one form or booking button. Generic homepages leak paid clicks.
Cut wasted spend
On Google, add negative keywords from the search terms report. On Meta, exclude areas you do not serve and refresh tired ads.
Reply within minutes
Speed does not change CPL, but it changes what each lead is worth.
Track to the sale
Connect leads to bookings and revenue, even in a spreadsheet. To see what outside help costs, read what a marketing agency costs.
Frequently asked questions
What is a good cost per lead?
A good cost per lead is one that leaves you profitable once leads become customers. Multiply customer value by gross margin, by the share of that margin you will spend on ads and by the percentage of leads that buy. The result is the most you can pay for a lead. Benchmarks, such as $66.69 on Google search and $27.39 on Meta lead ads in US data, are a reference point, not a target.
What is the average cost per lead on Google Ads and Facebook?
In US data published by LocaliQ and WordStream in 2026, the median cost per lead was $66.69 on Google search campaigns and $27.39 on Facebook lead campaigns. Industry figures vary widely, from about $30 for auto repair on Google to about $132 for legal services, and the two platforms bring in different kinds of lead.
Is a lower cost per lead always better?
No. A cheap lead that never books can cost more per customer than an expensive lead that buys. Compare channels on cost per booked customer, which combines cost per lead with the share of leads that become paying customers.
What is a good cost per lead in the UK?
We could not find a reliable published UK benchmark. The one UK-focused source we opened does not state its geographic scope and appears to convert dollar figures into pounds, so we do not use it. UK businesses are better off calculating their own limit and testing against it for a few weeks.
When can I trust my own cost per lead?
Your own numbers beat any benchmark. As a rough rule, wait for a few dozen leads before judging a campaign, and track how many became customers. Some Meta lead benchmarks here rest on as few as four campaigns, so use them cautiously.
Sources
- LocaliQ: Search advertising benchmarks (page updated June 1, 2026). Primary source for Google search cost per lead by industry. Gives no sample size, period or countries on the page
- WordStream: 2026 Google Ads benchmarks (updated September 16, 2026). Primary source for the Google methodology: 13,474 US search campaigns, April 2025 to March 2026, medians in US dollars. Industry figures match LocaliQ's
- LocaliQ: Facebook advertising benchmarks (updated September 23, 2026). Primary source for Meta lead-campaign cost per lead. No sample size on the page; its text contradicts its own table in one passage
- WordStream: 2026 Facebook ads benchmarks (updated September 15, 2026). Primary source for the Meta methodology: 452 US lead campaigns, at least four per subcategory, medians in US dollars
- PPC Land: Google Ads CPL drops for the first time since 2020 (June 29, 2026). Secondary write-up, used only for the year-on-year change ($70.11 to $66.69)
- PPC Land: Facebook lead costs hold at $27.39. Secondary write-up, used for the prior-year figure ($27.66) and the note that the medians do not reconcile
- Visionary Marketing: Google Ads benchmarks 2026 (May 2026). Checked for UK data; no geographic scope and very small samples, so we did not use its figures
- ZenABM: LinkedIn ads benchmarks (April 2026). One agency's B2B data, cited once as a rough reference; the page does not explain how cost per lead was calculated