Stop advertising to strangers

When you launch an ad campaign for your event, the normal move is to fill in a form telling Meta who your audience is: ages 18 to 35, a given region, interested in electronic music and festivals.

Notice what you just did. You told the platform what you think your audience looks like, leaning on categories the platform itself invented. Meanwhile, you have a list of several thousand people who already bought a ticket, with full names attached, that you’re not using.

The audience you describe isn’t the audience that buys from you

The interests you pick in an ads panel are an approximation. They’re built from what people consume and follow, and they’re reasonably useful as a starting point. But they have a structural limitation: they describe whoever resembles your idea of your audience, not whoever actually shows up at your venue.

Ask a promoter to describe their typical attendee, and they’re almost always half right. The half that’s wrong is usually the interesting one. It’s the people who come along with someone else, who discover the event late, who travel in from another city, who are ten years older than you’d assume. That group doesn’t fit the description — and it fills the room.

When you build your campaigns on interests alone, that invisible audience gets left out systematically. And because it’s left out, it never shows up in your data. And because it never shows up in your data, next season you go back to describing the same audience as always.

There’s an alternative, and you’ve already paid for it

Ad platforms let you do something far better: upload your own buyer list and work with it directly. These are called custom audiences, they’ve existed for years on Meta, Google and TikTok, and in live events they’re clearly underused.

Starting from a list of real buyers, you can do two things — and the second one is the important one.

Advertise to people who already know you. You don’t have to explain who you are or why they should trust you. This is your cheapest audience to convert and the fastest to decide. It works well for presales, for second-day pushes, and for winning back people who came two editions ago.

Ask the platform to find people like them. This is where the real value is. Instead of you describing your audience, you hand the platform a real sample and ask it to find people with similar behaviour. The platform cross-references thousands of signals you can’t even see, and it usually surfaces profiles you would never have picked by hand.

It’s the difference between telling someone “find me people who like rock music” and telling them “find me people similar to these four thousand who bought my ticket.”

Why this works especially well for live events

There’s a reason this technique performs better for live events than for most other sectors: your buyer list is exceptionally good.

Someone who buys a ticket hasn’t clicked an ad out of curiosity. They’ve paid money upfront to reserve a night of their life three months from now. That’s a far stronger intent signal than almost any other product carries.

You also have something other sectors don’t: repetition. People who’ve been to several editions are a small, extraordinarily well-defined group. If you feed the platform only those, the sample is cleaner and what it finds looks more like your real audience.

It’s counterintuitive but usually true: an audience of a thousand loyal buyers outperforms one of twenty thousand generic emails.

The three mistakes that ruin the result

Uploading your whole database unfiltered. If you mix real buyers with people who left their email in a giveaway, the signal gets muddy and the platform ends up looking for people similar to an average that doesn’t exist. A small, good list beats a large, messy one.

Only using it for retargeting. Many promoters upload their list and use it purely to re-target people who already bought. That’s fine, but it’s the less valuable half: the growth is in the lookalike audiences.

Leaving it frozen. An audience uploaded by hand in January is stale by March. It neither picks up new buyers nor drops people who’ve already bought the ticket you were trying to sell them. The list needs to update itself, or it ends up working against you.

Where to start this season

You don’t need to rebuild your entire ad strategy. One controlled test is enough to decide.

In your next campaign, set up two ad sets with the same budget, the same creative and the same message. One targeted at your usual interest-based audience. The other at a lookalike audience built from your buyers. Run them in parallel and compare.

It’s a cheap test, it takes no longer than the campaign itself, and it finally answers a question this industry has been settling by gut feeling for years. You can even set it up inside the campaign you already have running right now.

Tools like Nevent keep these audiences synced with your ticketing platform so they update themselves as sales come in. But you can run the first test with a manual export, and what it tells you will shape how you spend the rest of your budget.