Why nobody will give you a straight number
Type this question into Google and you will find pages confidently announcing that gyms pay a specific dollar amount per Facebook lead. Open three of them and the ranges barely overlap. None of them tell you what data they are drawn from, what counted as a lead, which country, which month, or what the offer was.
That is not a coincidence and it is not laziness. It is because “cost per lead” is not a standard unit of measurement. It is a fraction. The numerator is ad spend, which is at least well defined. The denominator is “leads”, and every advertiser gets to decide privately what a lead is. Change the definition and the number moves by a factor of five without a single thing changing inside the ad account.
So the honest answer to “what should a gym pay per Facebook lead” is: it depends on four things you control and three you don’t, and the range across real gyms is wide enough that the midpoint is useless to you. What follows is the machinery underneath it, because once you understand the machinery you can tell in about ten minutes whether your own number is good or bad — which is the only question you actually care about.
Three gyms, one campaign, three different “costs per lead”
Imagine three gym owners running the identical campaign, identical creative, identical budget, in identical towns. Each spends the same amount in a month. Each gets the same number of people to tap the ad and fill something in.
- Owner A counts a lead as anyone who submits the instant form on Facebook. Name, email, phone, three taps, never leaves the app.
- Owner B counts a lead as someone who clicked through to his website and completed a form there, including a question about their goal.
- Owner C counts a lead only once they have booked a time on the calendar for an intro session.
Owner A will report the lowest cost per lead of the three, by a distance. Owner C will report the highest. Owner C is not doing worse. He may well be signing more members from the same spend, because his denominator is made of people who have already committed to a Tuesday at 6pm. His cost per lead looks terrible on a spreadsheet and his cost per member is the best of the three.
Same campaign · same spendRelative, not data
Cost per lead, as each owner reports it: Owner A lowest, Owner B middle, Owner C highest. Cost per member, what actually matters, can flip it: Owner A highest, Owner B middle, Owner C lowest.
This is the single biggest reason published benchmarks contradict each other. Before you compare your number to anybody’s, write down your own definition. If an agency quotes you a cost per lead and does not immediately say what they are counting, the number means nothing.
The seven things that actually move your cost per lead
1. How many people the auction has to choose from
Meta sells impressions through an auction. Meta’s own description of it is that the winning ad is the one with the highest total value, made up of your bid, an estimate of how likely the person is to take the action you asked for, and a quality assessment of the ad itself — Meta publishes the components here. Two practical consequences for a gym.
First, your radius is your market. A gym in a dense suburb with two hundred thousand adults inside a fifteen-minute drive is bidding into a deep pool. A gym in a town of nine thousand is bidding into a puddle, and it will exhaust that puddle in weeks. The small-town gym is not being punished for doing anything wrong; there is simply less inventory, and the same people see the ad again and again.
Second, you are not only bidding against other gyms. You are bidding against everybody who wants the attention of a thirty-four-year-old within five miles of your door, which in late November includes every retailer in North America. This is why the same campaign gets more expensive in Q4 and cheaper in mid-January without you touching it.
2. What you are asking for in exchange
The bigger the ask, the fewer the leads, the higher the cost per lead — and usually the better the lead. A free trial class costs less per lead than a paid six-week programme. A form that asks for a phone number costs more than one that asks for an email. A form with a question about the person’s goal costs more than one without.
None of that is a problem to be fixed. It is a dial. Turning it toward “cheap” fills your CRM with people who tapped a button; turning it toward “expensive” fills your calendar with people who typed a sentence about why their back hurts. Pick deliberately, then stop comparing your number to gyms who picked the other end.
3. Whether the creative looks like your gym
The ads that work for gyms are, almost without exception, the ones that look like the actual building and the actual members. Phone footage of a 6am class. A coach talking to camera for forty seconds about who the programme is for. A photo of the rig with the chalk on it.
Stock imagery and template creative underperform for a structural reason as well as an aesthetic one: everybody running the same agency template is showing the same image to the same people in the same auction. The ad that looks like a real place within driving distance is the one that earns the click, and the click rate feeds straight back into the auction estimate that decides what you pay.
4. Budget size and the learning period
Meta’s delivery system needs a volume of conversion events before it stops guessing. A campaign running on a very small daily budget can sit in that guessing state more or less permanently, which means the cost per lead you observe in week one is noise, not signal. It is also why splitting a small budget across five ad sets is usually worse than putting all of it behind one.
The practical rule: do not draw conclusions from the first week, and do not draw conclusions from fewer than about thirty leads. With a small number of events, one good day changes your average enormously.
5. Frequency and audience fatigue
If you have shown the same three ads to the same forty thousand people for eighteen months, your cost per lead will have crept upward the whole time and you will have blamed the platform. Check frequency. If the same person has seen the ad ten times this month, you do not have a bidding problem, you have a creative refresh problem.
6. Where the lead lands
If you send traffic to a website, the page is part of your cost per lead. A slow page, a page that opens on a photo carousel instead of a form, a booking widget that asks for a password — each one quietly raises the price of every lead you get. This is why we treat the landing page as part of the ad spend, not as a separate project that happens later.
7. Season
Demand and price both move with the calendar. In the accounts we run, January behaves nothing like August, and the members who join in the January rush are the hardest ones to keep. September is the month we think is underrated: the back-to-routine crowd is less resolution-driven and, in our experience, tends to stick. Whatever benchmark you build for yourself, build it with the month written next to it, because a February number and an August number are not the same number.
The number that matters more
Cost per lead is a diagnostic. It is not a business metric. Your business metric is cost per member who is still paying you in six months, and the only way to get there is to walk the whole chain.
| Step | What you measure | Where it breaks |
|---|---|---|
| Spend | Dollars into Meta this month | Rarely |
| Leads | Name + phone, by your definition | Definition drift |
| Contacted | Reached a human within minutes | Very often |
| Booked | Has a time on the calendar | Often |
| Showed | Physically walked in | Often |
| Signed | Payment details on file | Sometimes |
| Retained | Still paying at month six | Almost never measured |
Two divisions matter. Spend divided by signed members is your cost per member. Spend divided by members still paying at month six is the number that tells you whether the marketing was an investment or a purchase of temporary noise. To work out what you can afford to pay for a member in the first place, run your numbers through the gym marketing budget calculator.
In our experience, a lot of the gyms that come to us saying “our leads are too expensive” do not have an ad problem. They have a gap between “lead” and “contacted” measured in hours or days. The ads generated the interest and nobody picked it up while it was warm. That is a cheaper problem to fix than a bidding problem, and it is why we build the follow-up system before we scale the ad budget. When Jose at MUV The Gym collected $4.2K in new memberships in his first forty days, the ads brought the enquiries in and the follow-up made sure none of them sat in an inbox overnight.
Build your own benchmark in thirty days
You do not need an industry figure. You need yours. Here is the shortest honest version.
- Write the definition. One sentence, taped to the wall. Everyone who touches the CRM uses the same one.
- Run one campaign, one offer, one audience, for thirty days. Do not change the creative in week two because a Tuesday was slow. You are collecting a measurement, not optimising yet.
- Record six numbers weekly: spend, leads, booked, showed, signed, and the median minutes between the lead arriving and a human speaking to them.
- At day thirty, do the three divisions: spend / leads, spend / showed, spend / signed.
- Set a calendar reminder for month six and count how many of those signed members are still on the billing run.
Now you have a benchmark that is actually about your town, your offer and your front desk. Every future month gets compared to that, not to a blog post. And when the number moves, you have five other numbers to tell you which link in the chain moved it.
When a cheap lead is the expensive one
A very low cost per lead is a warning as often as it is a win. The usual causes:
- The offer is too generous. A genuinely free month attracts people who wanted a free month. They convert at a rate that makes the cheap lead the most expensive thing in the account.
- The radius is too wide. Leads from forty minutes away are cheap because there are more of them. They do not show up on a Tuesday in February.
- The form is too easy. Pre-filled instant forms produce mistaps. Adding one manual field — a phone number the person has to type, or a single goal question — raises the price and raises the quality.
- The lead is not local at all. Check the area codes. If they are not yours, something is wrong with the geography or you have found a bot farm.
And the opposite: a high cost per lead is fine, sometimes correct, when the lead is a booked appointment with a real person who typed a real sentence. Judge the price against what the thing is, not against a number you read somewhere.
If an agency quotes you a number on a call
Three questions. What are you counting as a lead. From how many gyms and over what period. What was the cost per signed member in the same accounts. An honest answer to the third question is rare and it tells you almost everything about who you are talking to.
We don’t quote a cost per lead before we have seen your market, your offer and your close rate, because a number given before that is a guess wearing a suit. If you want to go deeper on how the ad side actually gets built, our gym Facebook ads page explains what we run and how we report on it.
If you already have thirty days of your own numbers and you want a second opinion on which link in the chain is leaking, book a free call and bring the spreadsheet. Thirty minutes, no card, and if the answer is “your ads are fine, fix the phone” we will tell you that.
