Lookalike Audiences for Gyms: How to Set Them Up in Meta

A lookalike audience lets Meta find people who resemble your best members, then shows your ads to them inside your local area. For a gym, the result depends on three things: the quality of the member list you feed it, the narrow percentage you start with, and whether you judge it on joins instead of clicks. This guide covers all three, step by step, plus the privacy rules most gym owners never check.

Enochmarketing

Reach More Ideal Local Members

Enoch Marketing helps CrossFit gyms build tailored Meta campaigns, lead funnels, and local marketing strategies to grow in their markets.

Explore gym marketing services

TL;DR

A lookalike audience is built from a source list (members, leads, trial signups). Meta then finds similar people. The list you feed it matters more than any setting.
Build the source from your best members, not everyone you have ever emailed. Clean the list first.
Start at 1% to 3%. Lookalikes can't be limited to a city, so set the radius or ZIP codes at the ad set level.
Judge it on trials booked and members joined, not click-through rate.
Customer data used for ads carries consent and privacy obligations in the US. Check them before you upload anything.

What a lookalike audience is, and why gyms use it
Choose the right source audience
Check the privacy rules before you upload
Create the lookalike audience step by step
Pick a percentage and start narrow
Launch the campaign
Our perspective: lookalikes are a data project, not a targeting trick
Measure results and compare against other audiences
Common mistakes gyms make with lookalikes
Where we can help
FAQ
Sources

What a lookalike audience is, and why gyms use it

A Meta lookalike audience is a group of people who share significant traits with your current customers. You hand Meta a source, such as a list of customer emails, and it finds others who look similar to that group.

For a gym, this solves a real problem. Interest targeting such as "fitness" or "CrossFit" reaches millions of people who will never drive to your box. A lookalike starts from people who actually walked in, paid, and stayed. That is a far better description of your buyer than any interest category.

It works best in a specific situation: you already have a base of members and leads, you want more of the same kind, and you are running ads on Facebook and Instagram. If you have no list at all yet, start with a broad local campaign and a strong offer. Our guide to Facebook ads for gyms covers that starting point.

What a lookalike audience is, and why gyms use it — overview diagram

Choose the right source audience

Everything downstream depends on this step. Meta learns from the list you give it, so a list full of one-time visitors, expired leads and wrong emails teaches it to find more of the same.

Which lists make a good source

Ranked roughly from strongest to weakest for most gyms:

Active, long-tenure members. People who joined and stayed. One gym software company suggests seeding with members who attend at least five times per month, because that behavior signals real interest.
Members who bought the offer you want more of. If your best revenue comes from coaching add-ons or unlimited memberships, seed with those buyers.
Trial signups who converted. These show who responds to your ads and then buys.
All leads. Useful only as a fallback, since most of them never joined.
Website visitors and social followers. The weakest signal. Many are friends, family and curious scrollers.

A lookalike of members who left after 60 days will find more people who leave after 60 days. If you can split your list by retention, do it.

How big the list needs to be

The floor is low. Shopify's instructions say the source must contain at least 100 people from the same country, while Meta recommends a source of 1,000 to 5,000 people. Most single-location gyms have fewer than 1,000 active members, so you will often sit between the two.

A practical middle path: one fitness-ads guide suggests testing a 1% to 3% lookalike from a member list of 200 or more people. If your list is smaller than that, combine members with converted trial signups and past members who stayed at least several months. Quality beats raw count, but a tiny, noisy list will give inconsistent results.

Clean the list before you upload it

Gym CRMs collect duplicates, old phone numbers and conflicting membership statuses. One CRM vendor's cleanup guide suggests checking data separately for completeness, uniqueness, consistency, timeliness, validity and accuracy. In plain terms:

Remove duplicate contacts.
Make sure "active" really means active, and "cancelled" is marked as cancelled.
Drop invalid emails and phone numbers instead of guessing at corrections.
Keep recent members. A list that is three years old describes a different gym.

This is an afternoon of work, and it is the cheapest improvement you can make to the campaign.

Check the privacy rules before you upload

Uploading member data to an ad platform is a use of that data for advertising. Treat it that way.

The Federal Trade Commission has told companies that when information is collected in a confidential context, using it for advertising requires affirmative express consent, given through a separate agreement with clear and conspicuous disclosure. Terms buried in a privacy policy do not meet that standard.

A gym is not a pharmacy, but the principle reaches you. In an FTC case, a health app was accused of sharing users' health information with advertisers like Facebook and Google without permission. The settlement prohibits sharing health data with third parties for advertising.

For your gym, that points to a few habits:

Upload contact details only. Do not include health screening answers, injury notes, body measurements or medical conditions in anything you send to an ad platform.
Say it at signup. Add a clear, separate line to your waiver or signup form explaining that contact details may be used to show you ads, with a way to opt out.
Honor opt-outs. Remove anyone who unsubscribed or asked not to be marketed to.
Ask a lawyer about your state. Privacy law differs between states, and your attorney can tell you what applies to your members.

This is general guidance, not legal advice. A short conversation with a business attorney is worth it before your first upload.

Create the lookalike audience step by step

The exact menus move around, but the flow has stayed consistent.

Upload your source. Create a Custom Audience from your cleaned member or lead list in Meta's audience tools, or use a pixel or conversion event if you track joins on your site.
Start the lookalike. In the campaign's audience step, choose Lookalike Audience, pick your Custom Audience (or a Page) as the source, select a country, and set a range from 0% to 10%.
Name it clearly. Something like "LAL 1% US - active members Oct" will save you confusion later.
Wait for it to populate. One Meta ads educator reports that population can take six to 24 hours before you can use it.
Attach it to an ad set. Select the lookalike in the ad set's audience field.

The local targeting catch

Here is the part that trips up most gym owners. When you create a lookalike, you choose a country, and cities and neighborhoods cannot be used as its location. That means your lookalike is built from people across the whole US.

You narrow it afterward. Geography is selected at the ad set level: apply a city, ZIP codes or a radius there. The same source reports that US radius targeting can be as tight as 1 mile.

So the workflow is: build a country-level lookalike, then restrict the ad set to the neighborhoods your members actually come from. Look at your member addresses first. If most live within a few miles, set a radius that fits, and consider ZIP codes if your area has natural barriers like a river or highway.

Country-level lookalike narrowed to local areas

Pick a percentage and start narrow

The percentage controls how closely the audience matches your source. Lookalike size can range from 1% to 10% of the target country's population, and smaller percentages are more similar to the source.

Because you will then cut the audience down to a local radius, a 1% country-wide lookalike is usually enough people to fill a metro area. Here is a simple testing plan:

Test	Source	Size	Why
A	Active members, 6+ months	1%	Tightest match, your baseline
B	Same source	2% to 3%	More reach, slightly looser match
C	Converted trial signups	1%	Checks whether recent buyers beat long-term members
D	Local interest or broad targeting	n/a	Your control, to see if lookalikes are worth it

One finance-for-sellers guide recommends starting with a 1% lookalike of high-value customers and expanding to 2% to 5% for scale, noting that larger percentages broaden reach while reducing similarity. In a local market, you will often hit audience saturation before you need to go past 3%.

Launch the campaign

Objective

Pick an objective that matches the result you want. Fitness ad guidance recommends a conversions objective tied to a trial sign-up rather than traffic or engagement. If you optimize for clicks, Meta finds people who click, which is not the same as people who book.

If your site tracking is not set up, a lead form can be a faster route, and our walkthrough of lead form ads for gyms shows how to run one. Either way, make sure the join or booking is counted. We explain how in conversion tracking for gyms.

Creative and offer

Audience is only one of the levers. The creative needs clear ad text and branded images that show your services. For a gym, that means real photos of your floor and your coaches, not stock images of models.

Run two or three creative options per ad set. The same fitness-ads guide suggests letting them run for at least seven days before judging performance, which gives Meta time to leave its learning phase and gives you enough data to compare.

Your offer matters as much as your targeting. A lookalike audience with a vague "join now" ad will lose to a broad audience with a clear first step. We cover this in offer testing for gym ads.

Our perspective: lookalikes are a data project, not a targeting trick

We work only with CrossFit gyms and fitness studios, and the pattern is consistent. Owners hear "use lookalikes" and expect a switch that cuts their cost per lead. Then they upload a messy export of every email the gym has ever collected, pick 5%, and conclude that lookalikes don't work.

The audience is rarely the problem. The usual causes are upstream and downstream of it:

Upstream: the source list is stale, mixed, or full of people who never joined.
Downstream: the ad leads to a slow website, nobody follows up on leads within minutes, or the gym can't tell which leads became members.

We treat the lookalike as one piece of a connected system: the offer, the ad, the landing page, the follow-up, and the tracking. If one link is weak, a better audience won't rescue it. That is why we like to audit the whole path to a booked trial before touching audience settings.

If you take one thing from this guide, make it this: feed Meta your best members, keep the targeting local, and measure the result on members, not clicks.

— Collin

Measure results and compare against other audiences

In Ads Manager's Insights, track frequency, amount spent, link clicks, CPM, CTR and CPC, and compare which audiences perform best. Those are useful diagnostics, but they are not the finish line.

The numbers that matter for a gym:

Cost per booked trial. Platform leads are not enough. Count the people who show up.
Cost per new member. Divide spend by members who joined from that audience.
Show rate and close rate. A cheap lead who never shows is expensive.
Frequency. If it climbs quickly in a small radius, your audience is saturating. Refresh creative or widen the lookalike.

To compare fairly, keep the offer, creative and budget the same across ad sets and change only the audience. Give each test enough time and spend to produce real data before you cut anything. For a deeper look at what to measure, see why track fitness ad performance and why track conversions in gym ads.

Refresh the source over time

Your membership changes, and so should your source. Re-export and re-upload a clean list every few months, and rebuild the lookalike from it. One educator reports that an actively targeted lookalike can update every three to seven days, but it only updates relative to the source you gave it. A stale source means a stale audience.

Common mistakes gyms make with lookalikes

Using every contact as the source. Past leads who never joined dilute the signal.
Going to 5% or 10% right away. Similarity drops as the percentage rises, and you are paying to reach people who barely resemble your members.
Expecting city-level lookalikes. Set geography at the ad set instead.
Judging after two days. Ads need time and enough spend to produce a meaningful read.
Skipping follow-up. The best audience still produces nothing if leads wait a day for a reply.
Uploading sensitive data. Contact details only, with consent.

Where we can help

If you would rather not build this yourself, we run it as part of our paid advertising for gyms, connected to your site, follow-up and tracking. Our monthly plans start at $1,500 per month for Get Found, with Fill the Floor at $2,500 per month and Growth Engine at $4,500 per month. You can compare them on our pricing page. For a wider picture of how Meta fits with the rest of your marketing, read our Meta ads guide for gyms.

FAQ

How many members do I need to build a lookalike audience?

Meta needs a source of at least 100 people from the same country, and it recommends 1,000 to 5,000, according to Shopify's instructions. Many gyms fall in between. If you have a small list, focus on quality: use your most engaged members and converted trial signups, and expect more variation in results.

Should I use a 1% or 10% lookalike?

Start at 1% to 3%. Smaller percentages match your source more closely, and larger ones reach more people with less similarity. Because you will limit the ad set to a local radius, you rarely need the larger sizes.

Can I target a lookalike audience to my city only?

Not at creation. You pick a country when you build it. You then apply city, ZIP code or radius targeting in the ad set, which narrows the lookalike to your service area.

Is it legal to upload my member list to Meta?

It depends on how you collected the data and what you told members. The FTC has said that using confidential information for advertising calls for clear, separate, affirmative consent. Upload contact details only, never health information, and confirm your state's rules with an attorney.

Do lookalike audiences work better than broad targeting?

Not always. Test them against a broad or local-interest control with the same offer and creative, and compare cost per booked trial and cost per new member. Let the data decide for your gym.