How I create lookalike audiences for music ads in 2024


Lookalike audiences are a powerful way to apply account-specific cold targeting to your ads.

Now, technically speaking, lookalike audiences are cold-targeted options akin to using similar artists, genres, and other affinities in your campaigns.

However, in my experience, lookalike audiences are generally better and more powerful.

Almost like a “warm” cold targeting, if that makes any sense at all.

And what’s more is, because these audiences are built upon your data, not some arbitrary targeting option that’s predetermined by Meta, they will grow and evolve as you move forward on your ad journey.

Becoming even better over time.

So let’s take a look at my favorite way to make lookalike audiences to promote music in 2024.

Custom audiences

Before we can build our first lookalike audience, we need to start with a custom audience.

This is because every lookalike audience is based on a custom audience as its source or “seed”.

Now, if you’re an existing subscriber, you’ll remember my breakdown of 6 custom audiences from last week.

I even did a YouTube video on it if that’s more your thing.

But to save ourselves a bit of time jumping around, let’s have a quick refresher on how this works.

First, we’ll take the source of our choice (e.g. Instagram engagement from the past year) and create a custom audience from that.

This will dump people who have previously engaged with our content into one group that we can use.

We’ll then take that custom audience and use it as the source for one or more lookalikes.

Easy.

Set up for success

Custom audiences pull people from everywhere—they’re not location-specific.

Lookalike audiences, however, will only pull users from the locations we specify during the setup process.

Now, if we’re creating a lookalike audience from one country, it’s as easy as dropping that country into the lookalike creation process and moving on with our day.

But sometimes, we want to target multiple countries.

And inputting all those countries manually can take time because, unfortunately, we can’t just copy and paste our countries into this field like we can at the ad set level during the campaign creation process.

So let’s make our lives easier by saving a group of countries for later use.

First, we’re going to head to that blue “Create audience” button in the top left corner of the audience manager and select “Saved audience”.

Note: We’re not actually creating a saved audience here, just using this menu to save our locations.

Now we can copy a list of countries like this one and paste it into the “Locations” field.

Next, we’ll be met with a little dialogue box that looks like this:

Select “Countries” from the dropdown, then select “Match locations”.

Once the list is generated, we can remove any locations we don’t want (I usually pull Aland Islands, Netherlands Antilles, United States Minor Outlying Islands, and US Virgin Islands from this specific list), click the box next to “Save this list to use it again”, then select “Add 13 locations”.

And finally, give the list a name (something like T1 for this list), then back out of everything to return to the audience manager’s main screen.

So now that we have our custom audience and our saved locations, we can finally create a lookalike audience.

My lookalike process

To start, we’re going to visit that blue “Create audience” button again in the top left of the screen but instead of selecting “Custom” or “Saved” audience, we’ll select “Lookalike audience”.

When the lookalike creation box appears, the first thing we’ll do is select our source.

Keeping with our “Instagram [365]” audience we touched on previously, we’ll start with that.

Note: To save yourself some time searching, simply type in the name of the audience you want to use as your seed and find it under the “Other sources” tab when it appears.

Next, we’re going to click the browse button under the “Select audience location” section, select “Saved locations”, then check the box next to our saved group of countries to have them automatically populate for us.

And once that’s done, all we have left to do is to choose the size of our audience at the bottom.

Now there are a lot of competing theories surrounding lookalike audience size.

My belief is that what you choose to do here will be largely dictated by your goals and your industry.

Because music is such a massive global product with broad market appeal (genre notwithstanding, of course), I believe wider is better for our use case.

I also understand that, just as with custom audiences, Meta knows more about who is in this audience than we do, meaning the system is likely to start showing our ads to people who are within that 1% range before moving down the line to 10%.

We are only helping ourselves in the long run by including as many people as we can.

So, for this last step, I like to select 10%.

This is going to give me the biggest audience with the most opportunities.

After that, select “Create audience” and you’re done.

More audiences to consider

So now that you understand the process, you can create as many lookalike options as you want.

My favorite combination is to create that “Instagram [365]” audience and then create 10% lookalikes in T1 countries, T2 countries, and T3 countries.

I generally lean into T1, but I like to have each of these just in case I want to go global with my targeting.

Now, obviously, I’ve created more options than just these three, but these are the ones I lean into the most.

But it doesn’t have to stop there.

You can even take each of the remaining five custom audiences we’ve previously made and create three, tier-based lookalike audiences for them.

Or you can make single-country-specific audiences like I’ve done to bring things in even further.

How you choose to use this tool is entirely up to you.

Don’t be afraid to mix, match, and test.

Because testing is the name of the game.

Whenever you're ready, there are four ways I can help you:

  1. Read the Newsletter: Read previous issues of The One Thing to learn at your own pace and upgrade your marketing knowledge for free.
  2. Book a Consultation: Schedule a one-on-one call with me to improve your marketing across paid advertising, social media, and more.
  3. The Spotify Traffic Accelerator: Join the hundreds of artists who have successfully learned to automate their growth on Spotify using paid ads on Instagram.
  4. Become a DuPree X Artist: Hire our team to manage your marketing across streaming platforms and social media so you can focus on what matters most—making music.

The One Thing

One high-leverage idea to scale your audience (and your business). Delivered every Tuesday.

Read more from The One Thing

Testing new ad creative is an effective way to maximize the efficacy of your Meta advertising campaigns. By continually introducing new creative options—photo, video, music, etc.—we can more readily find our “winners” and ensure they remain the best fit for our audience for an extended period of time. So let’s have a look at what it takes to run a creative split test, examine when and why we might do that, and dig through how we can best determine our winning ad(s) on the other side. When and...

Having multiple targeting options at your disposal is always a good thing, but sometimes you need to know which one works best. Enter: the split test. Just because one audience works in one context doesn’t mean a different one won’t be better in another. And in most ad campaigns, starting with more than one option is the best way to go about it. Fortunately, there’s a simple way to test multiple audiences at once, no matter the campaign type or your advertising goals on the other side of it....

The right targeting can make or break your ad campaigns to grow on Spotify, so it’s always good to have options. I am constantly tinkering with targeting options inside my own ad campaigns so we can take those winning combinations and apply them to campaigns for others with DuPree X. I also just like to have more than one setup I know I can depend on in the event the first option I try doesn’t work. So, after all of my testing, here are the four ad set targeting setups I lean into the most....