How do I calculate my break-even cost per lead?

Take the revenue a customer produces over the whole relationship, not just the first job. Multiply by your gross margin to get gross profit. Multiply that by your close rate on leads, expressed as a decimal. The result is your break-even cost per lead. A $2,880 gross profit per customer at a one-in-four close rate gives $720. Halve it for a working target, because the first version of any campaign underperforms.

Is a cheap cost per lead always better?

No, and it is the most common way small budgets get wasted. Cost per lead falls fastest when you loosen the offer and widen the audience, which also collapses the close rate. Fifty leads at $4 that never book are worse than eight leads at $25 where three become customers. The only figure that matters is cost per customer, and cost per lead is just one input to it.

What counts as a lead?

Decide once and write it down, because most confusion here is definitional. A useful definition is a named person who asked about the work and can be contacted: a form fill with a real phone number, a booked call, a direct message describing a job. A follower is not a lead. A click is not a lead. An email address collected in exchange for a discount code is a lead of a very weak kind, and should be counted separately.

How many leads do I need before the number means anything?

Enough that one extra close would not swing the figure much. At a one-in-four close rate, thirty to fifty leads gives you a usable read. Under ten leads you are looking at noise and should not adjust anything yet. This is the single most common error: killing a campaign at seven leads because the two that answered were tyre-kickers.

What should I do if my real ceiling is lower than what leads cost?

You have four levers and none of them is a cheaper ad. Raise the price, raise the close rate by qualifying and answering faster, increase what a customer is worth by adding repeat or contract work, or change to a channel where people arrive already looking. If none of those move, the honest conclusion is that paid acquisition does not work at your current unit economics, and that is worth knowing before you spend another $2,000 finding out slowly.

Does cost per lead differ between Meta and Google?

Yes, and predictably. Search costs more per click because the person is actively looking, so leads are fewer and closer to buying. Paid social costs less per click and produces more, weaker leads because you are interrupting people. Comparing the two on cost per lead alone will always flatter social. Compare them on cost per customer.

How long before I can trust my cost per lead?

Roughly two to three weeks of steady spend, or whatever produces thirty to fifty leads at your volume, plus the length of your sales cycle. If jobs take a month to close, a two-week-old campaign has leads that have not had time to become customers, and reading it as a failure is premature.

What is a good cost per lead for a small business?

Aug 22, 2026 · 15 min read

There is no benchmark worth copying. The only honest answer comes from your own gross profit and close rate, and the arithmetic takes about twenty minutes.

A good cost per lead is any number below the ceiling your own arithmetic sets, and that ceiling is specific to your business. Take what a customer is worth in gross profit, multiply by the share of leads you actually close, and you have the most you can pay for a lead and still break even. Half of that is a healthy target. A bookkeeper closing one lead in four into a $400 monthly retainer can afford more than twenty times what a landscaper closing one in three into a single $180 mowing job can. No industry benchmark can tell you either number.

Key takeaways

There is no industry benchmark worth copying. The ceiling comes from your gross profit and your close rate.
The formula is gross profit per customer multiplied by your close rate on leads. That product is break-even.
Target roughly half of break-even, so a bad month does not put you underwater.
A lead, a qualified lead, and a customer are three different things, and mixing them makes every figure meaningless.
Cheap leads that never close are more expensive than costly leads that do. Judge on cost per customer.
If the honest ceiling is unaffordable, the fix is price, close rate, or repeat value — not a cheaper ad.

Search for this question and you will find a table of averages by industry: so many dollars per lead in home services, so many in fitness, so many in professional services. Those numbers are averages across businesses with different prices, different margins, different close rates, and wildly different amounts of repeat work. A national franchise with a call center and a two-truck landscaping outfit both sell lawn care. They cannot possibly share a good cost per lead.

The workable answer takes about twenty minutes and a calculator, and it requires no data you do not already have. You need three numbers you can estimate from last year: what a customer is worth over the whole relationship, what share of that is gross profit, and how many enquiries turn into paid work. Everything below is either how to get those three right or what to do when the answer they produce is uncomfortable.

The number you want is a ceiling, not a benchmark

A benchmark tells you what other people paid. A ceiling tells you what you can afford. Only one of those changes a decision. When someone asks whether $38 per lead is good, the question cannot be answered without knowing what a lead is worth to them, and once you know that, the benchmark becomes irrelevant. Thirty-eight dollars is a bargain for a roofing company and ruinous for someone selling $22 candles.

The ceiling is also the only version that survives changing conditions. Ad costs rise, a competitor starts bidding on your audience, a season ends. Against a benchmark those all feel like failures. Against your own ceiling they are numbers moving toward a line you drew yourself, and you know how much room is left.

Deriving it also forces you to confront numbers most owners avoid: the real margin after materials and subcontractors, the real close rate rather than the flattering one. Owners who do this often discover the acquisition problem was never the ad. It was a 22 percent margin on a job they thought made 45.

The arithmetic, worked out loud

Three numbers, multiplied. Nothing about this is sophisticated, and the lack of sophistication is the point — a formula you cannot recompute in your head on a job site will not get used.

A landscaping example

Say you mow lawns. The average one-off job bills $180. After fuel, blade wear, dump fees, and the hours of the person on the mower, roughly half of that is gross profit, so $90. Of every three people who call for a quote, you win one. So one lead is worth, on average, one third of $90, which is $30. That is the break-even cost per lead. Pay $30 and you have converted advertising money into an equal amount of gross profit and taken home nothing.

Now change one thing. Instead of a one-off cut, you sell a seasonal contract: $180 per month for seven months, which is $1,260 of revenue and $630 of gross profit. The close rate stays at one in three. The break-even cost per lead is now $210. Same trucks, same crew, same phone, same ad — and seven times the amount you are allowed to spend to make the phone ring, purely because of what happens after the first job.

That comparison explains most of the confusion around this question. Two landscapers can look at the same $60 lead, and one is losing $30 while the other is making $150. Neither is wrong. They are running different businesses.

The same arithmetic in one line

Gross profit per customer, multiplied by close rate on leads, equals break-even cost per lead. Then take a third to a half of that figure as the number you aim for. The discount is not conservatism for its own sake. Your first campaign will underperform, cold leads close worse than referrals, and a business that only breaks even on acquisition has nothing left to grow with.

Illustrative ceilings across four business shapes
Business shape	Gross profit per customer, and close rate	Break-even cost per lead
Landscaper, one-off mow at $180	$90 profit at a 50 percent margin. Closes one lead in three.	About $30
Landscaper, seven-month seasonal contract	$630 profit on $1,260 of revenue. Closes one lead in three.	About $210
Bookkeeping practice, $400 monthly retainer kept a year	$2,880 profit on $4,800 at a 60 percent margin. Closes one lead in four.	About $720
Local gym, $79 monthly membership kept nine months	$569 profit at an 80 percent margin. Closes one trial request in five.	About $114
Shopify store, $48 average order, email signup as the lead	$22 profit at a 45 percent margin. One in twenty subscribers buys.	About $1.10

Notice the last row. A store collecting email addresses is playing a different game, where a lead has to cost about a dollar, which is why those businesses live on organic reach and repeat purchase. If you sell a $48 product and someone tells you a good lead costs $30, they are describing a business that is not yours.

Break-even cost per lead: The gross profit an average customer produces over the whole relationship, multiplied by the share of leads that become customers. Above this number, advertising loses money on every lead; below it, the gap is your margin on acquisition.

A lead, a qualified lead, and a customer are three different things

Most arguments about cost per lead are actually arguments about definitions. An advertising platform will happily report a $6 lead while your bank balance disagrees, because the platform is counting form submissions and you are thinking about people who answered the phone. Both counts are correct. They measure different objects.

A lead. A named person who asked about the work and can be contacted. A form fill with a real number, a booked call, a direct message describing a job. Not a click, not a follower, not an impression.
A qualified lead. A lead who has the problem, is in your service area, can afford the price, and is deciding now rather than in eight months. This is the number your close rate should be measured against.
A customer. Someone who paid. The only figure that pays wages, and the only one worth optimizing a campaign toward.

Track all three or the arithmetic silently breaks. If a campaign produces forty leads and only ten are in your service area, your real cost per qualified lead is four times what the dashboard says. That is a targeting problem showing up in the wrong column, and it is fixable in an afternoon once you can see it. The mechanics of getting the qualification into the ad itself are in how to write ad copy for a small business, where naming the price range and the area in the creative does most of the filtering for you.

A lead is not an achievement. It is a bet with a price attached, and the price is only good relative to what the bet pays.

Why cheap leads are often the expensive ones

Cost per lead is the easiest marketing number to improve and one of the easiest to improve destructively. Widen the audience, soften the offer, remove the price from the ad, replace a phone number with a one-field form, and the cost per lead will drop by half. So will the close rate, usually by more. The dashboard looks better and the business gets worse.

Illustrative: the same $500 spent four ways
Lead source	Leads and close rate	Cost per customer
Broad audience, vague offer, discount hook	100 leads at $5. One in fifty becomes a customer.	$250 per customer
Narrow audience, the exact job named, price range shown	40 leads at $12.50. One in eight becomes a customer.	$100 per customer
Purchased contact list	250 leads at $2. One in two hundred becomes a customer.	$400 per customer
Referral partner or repeat enquiry	8 leads at $62.50. One in two becomes a customer.	$125 per customer

The best cost per lead in that table produces the second worst cost per customer. The worst cost per lead produces a perfectly respectable one. If you had been managing to cost per lead, you would have shut down the campaign that was working and doubled the one that was not. This is not hypothetical arithmetic; it is the single most common way a small advertising budget gets destroyed, and it is why why my Facebook ads are not working spends as much time on lead quality as on the ads themselves.

Cheap leads that do not close still consume phone calls, site visits, and quotes written at nine in the evening. If your own labour is scarce, that cost appears in no advertising report.

One-off jobs and repeat customers are not the same business

The largest single swing in your ceiling comes from what happens after the first transaction, and it is the variable most owners leave out. A one-off job gives you one margin to pay for the lead. A customer who returns monthly gives you twelve, or thirty-six, and the lead price you can justify moves accordingly.

Be honest about retention rather than optimistic. If you have run for a year, count how many customers from last spring bought again this spring. If the answer is a quarter of them, the average customer is worth roughly one and a quarter jobs, not the five you would like to assume. Multiplying by a hoped-for lifetime value is how owners talk themselves into unsustainable lead prices.

If you do not have a year of history, use the first job only. That gives you a conservative ceiling you can act on immediately, and every repeat customer after that becomes upside rather than a load-bearing assumption. Service businesses in particular tend to find that the first ceiling is uncomfortably low and that the fix is structural — converting one-off work into a maintenance arrangement — rather than promotional. How to get clients for a new service business works through that conversion in more detail.

The close rate is the number most owners guess wrong

People remember the jobs they won and forget the quotes that went unanswered. Ask an owner their close rate and you will usually hear a number ten to twenty points too high, which inflates the ceiling and licenses overspending. It is worth reconstructing rather than recalling.

Count from enquiries, not from quotes sent. The enquiries you never followed up on are still leads you paid for. Dropping them from the denominator is the most flattering error available.
Separate cold leads from referrals. A referral closes at a much higher rate because trust arrived with it. Blending the two produces a close rate that is true of neither and will make paid advertising look affordable when it is not.
Count the ones that went quiet as losses. A quote with no answer for three weeks is a loss. Leaving it as pending forever is how a 30 percent close rate reports itself as 60.
Measure response time alongside it. Close rate is partly a function of how fast you reply. If the phone goes unanswered until evening, some share of your cost per lead is being spent on leads that hire someone else by lunchtime.

That last point is the cheapest improvement in this post. Raising your close rate from one in five to one in four raises your allowable cost per lead by 25 percent without touching price, margin, or advertising. Answering the phone is a marketing channel.

How to work out your own number in one sitting

This takes twenty minutes with last year’s invoices and a calculator. Do it once, write the result on the wall, and revisit it when your prices change.

Add up what an average customer paid you across the whole relationship. Not the biggest job. The average, including any repeat work, over however long they stayed.
Subtract the direct costs of delivering it. Materials, subcontractors, fuel, payment fees, and the hours of whoever did the work. What remains is gross profit. Do not subtract rent or insurance; those are not per-customer costs.
Reconstruct your close rate from records, not memory. Count every enquiry from one representative month and count how many became paid work. Express it as a decimal.
Multiply gross profit by the close rate. That is your break-even cost per lead. Say it out loud, because the number is often surprising in both directions.
Take one third to one half of it as your target. The gap between target and break-even is your margin on acquisition and your room for a bad month.
Write down the number of leads that will make the result readable. Thirty to fifty at a one-in-four close rate. Decide this before you launch, so you cannot quietly move the line later.
Run one narrow campaign for a fixed window and compare against the target, not against a benchmark. The only comparison that matters is with your own ceiling.

On budget sizing for that first window, how much a small business should spend on ads covers what a readable test actually costs, and how to run your first Meta ads campaign covers the mechanics of getting it live without wasting the first week on account setup.

What to do when the honest ceiling is unaffordable

This is the outcome nobody plans for and many owners reach. You work out that a lead is worth $18 to you, and leads in your category cost $45. That is not a reason to give up on advertising forever, but it is a reason to stop buying leads today. You have four levers, and adjusting the ad is not one of them.

Raise the price. The fastest lever and the most resisted. A 20 percent price increase on a 50 percent margin raises gross profit by 40 percent, and therefore raises your ceiling by 40 percent. Most small businesses are underpriced relative to their local market and have never tested otherwise.
Raise the close rate. Answer faster, qualify harder in the ad, follow up twice instead of never. Doubling the close rate doubles the ceiling and costs nothing but discipline.
Increase what a customer is worth. Add a second service, a maintenance plan, a contract instead of a callout. This is the structural fix, and it changes the ceiling more than anything else on this list.
Change channel rather than creative. If interruption advertising cannot work at your unit economics, a search-intent channel or an owned distribution channel may. That is a different question, and where are my customers online is where to take it.

What you should not do is keep the price, keep the close rate, keep the one-off job, and hope advertising gets cheaper. It does not. The auction is competitive and your competitors are learning too. A business that only works at last year’s ad prices has a countdown on it.

The first month of a new channel will always look bad

There is a timing trap in this metric that kills otherwise viable campaigns. Cost per lead is highest at the beginning, for two reasons that have nothing to do with whether the channel works. The platform is still learning who to show the ad to, and you are still learning which of your angles lands. Judging week one against your target is like judging a new hire on their first afternoon.

The other half of the timing problem is your own sales cycle. If your average job closes eleven days after the first call, then a campaign that has been live for twelve days has leads that have physically not had time to become customers. Owners regularly shut off a campaign at day ten, then get three bookings in the following fortnight from leads they already paid for, and never connect the two events. Decide the window in advance, and make it at least your sales cycle plus a week. How to read ad test results covers what a readable window looks like and which movements are noise.

When cost per lead is the wrong thing to be measuring

Sometimes the metric itself is a distraction. If people are arriving and not enquiring at all, you do not have a cost per lead problem; you have a conversion problem, and no amount of cheaper traffic fixes it. Traffic but no sales deals with that case directly. If nobody clicks in the first place, the creative or the audience is wrong and the lead figure is meaningless because the denominator is empty.

The metric is also wrong while you are still testing whether anyone wants the thing. In a demand test the point is not to buy leads efficiently but to find out whether cold strangers react to a claim at all, and that cost should be judged against the cost of building the wrong thing for six months. Confusing the two makes owners optimize a test that was supposed to be a question.

For a genuinely local business the number can mislead in a third way, because the population that could ever buy is small enough that you exhaust it and frequency climbs. How to market a local business online covers what to watch when your entire market is four postcodes.

What to do with the number once you have it

The value of the ceiling is that it converts a vague worry into a decision rule. When a channel produces leads under your target, you spend more. When it produces leads between target and break-even, you improve the close rate or the offer before spending more. When it produces leads above break-even for a full window with enough volume to be readable, you stop, and you stop without a discussion, because the number was agreed before anyone had feelings about it.

Review it twice a year and whenever your prices move. A price increase raises the ceiling immediately; a supplier squeeze lowers it. Owners who set this once end up managing this year’s advertising against a business that no longer exists. And if results keep coming back above break-even across several genuinely different attempts, the question stops being about lead cost and becomes a question about the idea, which is what when to quit a business idea is for.

How much should a small business spend on ads — what a readable test costs before you optimize anything
How to read ad test results — which movements in the numbers are real and which are noise
Traffic but no sales — when the leads arrive and nothing converts
Where are my customers online — if the answer is that you are buying leads in the wrong place
How to write ad copy for a small business — qualifying in the creative so fewer leads waste your evening

Frequently asked questions

What is a good cost per lead for a small business?: A good cost per lead is any number comfortably below your own break-even ceiling. Work out the gross profit a customer produces, multiply it by the share of leads you close, and that product is the most you can pay for a lead without losing money. Aim for roughly half of it. Because gross profit and close rate vary enormously between businesses, a bookkeeper on a monthly retainer can afford twenty times what a landscaper doing one-off mowing jobs can, and no published industry average tells you which of those you are.
How do I calculate my break-even cost per lead?: Take the revenue a customer produces over the whole relationship, not just the first job. Multiply by your gross margin to get gross profit. Multiply that by your close rate on leads, expressed as a decimal. The result is your break-even cost per lead. A $2,880 gross profit per customer at a one-in-four close rate gives $720. Halve it for a working target, because the first version of any campaign underperforms.
Is a cheap cost per lead always better?: No, and it is the most common way small budgets get wasted. Cost per lead falls fastest when you loosen the offer and widen the audience, which also collapses the close rate. Fifty leads at $4 that never book are worse than eight leads at $25 where three become customers. The only figure that matters is cost per customer, and cost per lead is just one input to it.
What counts as a lead?: Decide once and write it down, because most confusion here is definitional. A useful definition is a named person who asked about the work and can be contacted: a form fill with a real phone number, a booked call, a direct message describing a job. A follower is not a lead. A click is not a lead. An email address collected in exchange for a discount code is a lead of a very weak kind, and should be counted separately.
How many leads do I need before the number means anything?: Enough that one extra close would not swing the figure much. At a one-in-four close rate, thirty to fifty leads gives you a usable read. Under ten leads you are looking at noise and should not adjust anything yet. This is the single most common error: killing a campaign at seven leads because the two that answered were tyre-kickers.
What should I do if my real ceiling is lower than what leads cost?: You have four levers and none of them is a cheaper ad. Raise the price, raise the close rate by qualifying and answering faster, increase what a customer is worth by adding repeat or contract work, or change to a channel where people arrive already looking. If none of those move, the honest conclusion is that paid acquisition does not work at your current unit economics, and that is worth knowing before you spend another $2,000 finding out slowly.
Does cost per lead differ between Meta and Google?: Yes, and predictably. Search costs more per click because the person is actively looking, so leads are fewer and closer to buying. Paid social costs less per click and produces more, weaker leads because you are interrupting people. Comparing the two on cost per lead alone will always flatter social. Compare them on cost per customer.
How long before I can trust my cost per lead?: Roughly two to three weeks of steady spend, or whatever produces thirty to fifty leads at your volume, plus the length of your sales cycle. If jobs take a month to close, a two-week-old campaign has leads that have not had time to become customers, and reading it as a failure is premature.