Meta ads that work for a local business

Late February is when the spring plan gets written. In a med spa, a clinic or a salon that means the run up to prom, weddings and the first warm weekend, and it is the point in the year when somebody suggests putting money into Instagram.

The request usually arrives as “we need more reach”. That is the wrong thing to buy, and it is the most expensive thing to buy badly.

Here is how we spend a local budget on Meta, and why very little of it goes on finding strangers.

Prospecting is a volume game and you do not have volume

An advertiser selling something shippable can prospect all day. The whole country is addressable, the system has thousands of conversions a week to learn from, and a bad week is a rounding error.

A local business has a drive time instead. Inside it there is a modest number of adults who could plausibly book, and only a fraction of those are in the market this month. Three things follow.

• The system has too little to learn from. Automated campaign types need conversions to optimize against. A clinic booking a handful of consultations a week does not produce them quickly enough for cold targeting to get clever, so the budget goes on finding that out.
• Frequency climbs fast. A small audience sees the same creative again and again inside a fortnight. The reach number on the screen keeps rising while the same people get shown the same ad, and the comments sour before the bookings arrive.
• Interest targeting is not what it was. The detailed options advertisers used to build cold audiences from have been cut back for years, and anything touching health or personal attributes is restricted, for good reason. Guessing at a cold audience is now mostly guessing.

That does not mean Meta fails for a local business. It means the money works at a different point in the process.

The four audiences worth building

We run local accounts with most of the spend behind people who have already done something, and a small patient slice looking for new ones. In the order they pay:

1. Your own list, with consent. Past customers and patients, uploaded properly and refreshed. The cheapest bookings on the account come from people who already paid you once and forgot you existed.
2. Site visitors, split by page. Somebody who read the page about the treatment is a different person from somebody who read the blog. Bid on them differently and write to them differently.
3. Video viewers and profile engagers. Anyone who watched most of a video or interacted with the account in the last month or two. It costs nothing to build and it is the audience most accounts ignore.
4. A lookalike built from booked work, not from leads. Seed it with the people who turned up and paid, which means your CRM has to know who those are. A lookalike off raw form fills teaches the system to find you more form fills.

Prospecting gets what is left, aimed narrowly: the area you actually serve, one offer, one piece of creative you would be content to see four times.

Offer led, because a local ad has to ask for something

The creative that works here is not the creative that wins awards.

• One offer, in the first line. A consultation, a package with an end date, the thing people ring up about anyway. Brand awareness in a small market is a donation.
• Real faces and real rooms. Phone footage of the practitioner explaining what actually happens in the appointment beats the polished version almost every time, because it answers the question a nervous person is really asking.
• Built for sound off. Captions, and the offer on screen in the first two seconds.
• Refreshed every few weeks. Fatigue arrives quickly in a small audience. Shoot a batch in one afternoon rather than one ad at a time.
• Careful with before and after. Health and beauty creative is held to stricter rules than most categories, and an account restricted in March takes the spring with it.

The follow-up is the campaign

The most common reason a local Meta budget looks like it failed is that nothing happened after the form.

A lead from a feed is colder than a lead from search. Somebody was looking at pictures of a friend’s holiday and stopped to ask about a treatment. The interest is real and it evaporates in about an hour.

What we hold accounts to:

• The lead lands in the CRM the moment it is submitted, not in a file somebody downloads on Friday.
• A text goes out inside a minute, from the business number, asking one question rather than announcing anything.
• Five attempts across the first week, text and call mixed, then a slower nurture rather than silence.
• Every lead gets an outcome written down: booked, not booked, wrong fit. Without that the lookalike above is built on nothing.

The same budget with this in place and without it produces completely different numbers, and the ad account looks identical in both cases.

What to do this month

• Check the pixel is still firing and your customer list has been uploaded this quarter, before you spend a dollar on spring.
• Build the four audiences this week, even if you only advertise to two of them at first.
• Move the majority of the budget behind people who have already visited or bought, and cap cold prospecting at an amount you would be relaxed about losing.
• Shoot four short vertical videos in one afternoon: the room, the practitioner, the appointment explained end to end, and one customer speaking for themselves.
• Put the speed to lead text live before the first ad runs, then submit your own form from your own phone and time the reply.
• Write one spring offer with an end date, and keep it off every other channel so you can tell what it did.

Meta is the channel local businesses most often judge on the wrong number. Reach is cheap and attention is not, and the account that wins is the one where the follow-up is built as carefully as the ads. That is why our Meta ads work sits inside the CRM rather than beside it, and why the first month on a med spa account usually goes on the audiences and the follow-up before anybody touches the creative.