Someone asks what a lead should cost in your line of work. The honest answer starts with a range, because the average cost per lead by industry runs from about $27 to about $132 on Google search, and both ends are correct for the businesses in them.
Here are the current figures, where they come from, and the specific ways people misread them. The short version: use them to check you are in a sensible postcode, never as a target.
This is for you if you are budgeting for paid search and want to know whether your numbers are normal.
The average cost per lead by industry, 2026 figures
These come from WordStream’s 2026 benchmark study of 13,474 US search campaigns running April 2025 to March 2026. It is the best-documented dataset in this space, because it publishes both its sample size and its date range, which most do not.
| Industry | Cost per lead |
|---|---|
| Attorneys & legal services | $131.63 |
| Furniture | $106.70 |
| Real estate | $102.51 |
| Business services | $93.69 |
| Home & home improvement | $90.92 |
| Restaurants & food | $30.57 |
| Automotive repair, service & parts | $29.96 |
| Arts & entertainment | $26.84 |
| All industries | $66.69 |
The all-industry average is $66.69, with an average conversion rate of 8.18%.
Why the average cost per lead by industry varies fivefold
Cost follows value. It is that simple, and it explains almost the whole table.
A legal firm winning one case worth thousands can pay $131 for an enquiry and be delighted. A restaurant cannot, because a booking is worth a few pounds of margin. Advertisers bid up to what a customer is worth, so expensive industries are expensive because the leads are valuable, not because those advertisers are careless.
The second driver is how considered the purchase is. Choosing a solicitor takes weeks and several enquiries, so more people are in the market comparing, and each of them costs money to reach. Booking a table takes ninety seconds.
Read the table that way and it stops looking like a league table of competence. It is a map of what a customer is worth in each market.
The thing that changed this year
Search got cheaper, which has not happened for a while.
WordStream describes 2026 as “the first time in five years” that the overall average cost per lead has gone down, and reports that conversion rate increased for 87% of industries. That is a broad improvement rather than one category pulling the average.
Meta moved the opposite way over the same period. Its own quarterly filing reports average price per ad up 12% year over year for the quarter ending 30 June 2026, with impressions up 14%.
If you run both channels, that divergence is worth more to your budget than any industry average. One of your two platforms got cheaper this year and the other did not.
What the average cost per lead by industry cannot tell you
Four things, and each one has cost somebody money.
It does not know your margin. Two home improvement firms at $90.92 a lead, one keeping 40% and one keeping 8%, are having completely different experiences.
It does not know your close rate. Doubling how many enquiries you convert halves what each customer costs you, without touching the ad account.
It does not know your geography. National averages flatten enormous local variation, and a competitive city can cost several times the average for the same search.
It does not know your definition of a lead. If the businesses in that dataset count newsletter signups and you count booked consultations, you are not measuring the same thing at all.
How to use these figures without being misled
Use them for three checks and nothing else.
First, a sanity check on scale. If your industry sits near $90 and you are paying $400, something is wrong and it is worth investigating. If you are paying $30, check your lead quality before you celebrate.
Second, a viability check before you start. If your ceiling is $40 and your industry averages $90, paid search may not work for you yet, and knowing that before you spend three months proving it is genuinely valuable.
Third, a negotiation check. When an agency promises you leads well below your industry average, ask what they are counting as a lead. That gap usually lives in the definition rather than in their skill.
How to place yourself when your industry is not listed
Most businesses do not appear cleanly in any published category, and forcing yourself into the wrong row is worse than using the overall average.
Ignore the label and match on two things instead. What is one sale worth to you in profit, and how long does someone take to decide? A specialist B2B consultancy behaves like legal services, whatever its official category, because both sell considered purchases worth thousands. A takeaway behaves like restaurants even if it calls itself retail.
Geography matters at least as much. These are US national averages, and a competitive city can run several times the figure shown for the same search. If you advertise in one metro area, expect to sit above the national line rather than on it.
Then there is your own definition. If the advertisers in that dataset counted newsletter signups while you count booked consultations, you are not measuring the same event, and no adjustment fixes that.
Where my own accounts have landed
My range across live accounts is wider than the table, in both directions.
The cheapest was $2.68 a lead across 5,599 leads for a US auto transport company on Meta, which is a fraction of any published figure. The most expensive was $91.18 across 906 leads for a US automobile accessories business on Google, which sits above the all-industry average and was still the right price for a market where one order is worth thousands.
Neither number would look sensible on a benchmark chart. Both made the client money, which is the only test that counts. You can see how those accounts were put together if you want the working rather than the headline.
What to change this week
Three steps.
Find the row closest to your business on two things, what one sale is worth and how considered the purchase is, rather than on the category label. Compare your current figure against it, and treat any gap as a question rather than a verdict. Then work out your own ceiling from profit and close rate, and use that as your actual target from now on.
The wider piece on judging this metric covers where averages fit, why your close rate sets the real target covers how to build one, and the difference between lead cost and customer cost covers the figure your finance team actually cares about. If you want your own numbers read against the market, send them over.