how do facebook lookalike audiences work?
If you own an online business such as a dropshipping store and you’re wanting to seriously laser target potential customers, then learning about Facebook’s Lookalike tool should be an interest of yours. With this tool, you’ll be able to easily gain new customers that share similar interests to your previous ones. But how do they work?
On the creation of a Facebook Lookalike audience, you’ll be able to select a custom audience from data that is already held on your pixel, or alternatively you’ll be able to import information from other sources such as a mobile app. Once Facebook has this information, it’ll identify the audience and post your campaign in front of people that share similarities with your current customers (for example, hobbies, location, qualities, etc).
Throughout this article I will be discussing the importance of a lookalike audience, what custom audiences are available to use and tips on making your first lookalike audience campaign successful.
are facebook lookalike audiences important?
If you’re looking at building your brand by tapping into new audiences, then it’s most definitely important. You may think you know your perfect audience. But trust me, Facebook’s sophisticated algorithm knows a ton more people that are interested in what you’re advertising. You just don’t know it.
The creation of lookalikes audiences has been a game-changer and something most advertising agencies can’t go without.
However, you’ll only be able to create a Lookalike audience from a pixel that has already gathered data from potential customers.
The more information you have about your ideal customer through your pixel, the better your results will be from your lookalike.
what custom audiences can create a lookalike?
I’ve mentioned to create a Lookalike audience you must have data. This “data” is split into several custom audiences automatically by Facebook’s algorithm. It’s important to know this, so you can try and test each individual audience to see which gives you the best results.
1. website custom audience
A website custom audience is the easiest and probably the most used.
Once you’ve embedded your pixel within your website, it’ll start creating this custom audience from day one.
When you’ve inserted your pixel, it’ll fire each time a customer loads onto a certain webpage within your website (the page depends on how you set it up).
This is typically the “thank you” page after checkout or the “cart” page.
2. Customer List Audience
I’ve mentioned that you can import audiences if you have previous data that isn’t on your pixel. If you have access to your customer’s telephone numbers, email addresses, or Facebook IDs (highly unlikely).
You’ll be able to manually import them and Facebook will try to match this information up with users on their platform.
I’ve done this before as I used to advertise on Snapchat and wanted to make the transfer to Facebook, in my own experience Facebook matched around 55-65% of my customers.
3. app activity audience
f you’re not selling products and are advertising a game or service that requires an app download.
Then you’re able to create an audience based on the actions they’ve taken on your mobile application.
4. facebook audience
This is an interesting one as it’s based on the engagement of your videos, posts, pages, groups, etc.
For instance, you’ll be able to search for a lookalike audience that has watched a certain video over a set amount of time.
This also included engagement from their sister company, Instagram, allowing you to get the best possible data.
Normally, they’ll only hold this information for around a year so the data source will be fairly relevant and should have enough information to perform effectively.
creating successful Facebook lookalike audiences
The information I could give you on this sector could be pages and pages long, so I’ve tried to narrow down the most important tips I could think of.
1. quality data
For your lookalike audience, you want to choose a custom audience that has proven to be good data.
What do I mean by this? All data is good data, right? This isn’t the case when it comes to a lookalike audience and you should look at it like this. If 1,000 people walk into your store on a Monday and walk out again, then on Tuesday 100 people walk in and 5 buy something, something, you’re going to want to target similar people to those 5 that purchased something, rather than the 1,000 non-buyers on Monday.
You can easily avoid this but narrowing down your lookalike audience, instead of having it broad.
Instead of selecting a lookalike based on visitors to your website, you may want to go further down the purchasing chain. You can opt into targeting a lookalike that has purchased something or added an item to cart.
If you’re importing data from either an email list, mobile app or another advertising platform the same rules apply.
Don’t import all of your data, it may seem like a great idea but what you really want to do is import the data that holds value.
2. test every custom audience
If you’re new to digital marketing you may not know the secret to a well-performing campaign.
Let me tell you, it’s down to testing- a lot of it. The same applies to your lookalike audiences, test until you genuinely find your best-performing audience and expand from there.
Here are some custom audiences that I commonly see people overlook:
3. Video Audience
When launching a video-based campaign, I always get excited when it’s time to make a lookalike audience that is based on watch time.
I’ve seen great figures from this and it’s most definitely something you should test.
4. Newsletter subscribers
If someone has already subscribed to your newsletter, that tells me that they’re interested in what I’m selling but just not ready to make the purchase.
Targeting an audience that shares similar interests to people that are already intrigued can work great.
5. Adding a product to a wishlist
Doing a lookalike audience based on your products being added to a wish list can be powerful in some cases.
6. never use old data
The main reason for people’s lookalike audiences not performing to the ability they should is down to old data.
If you want great results, I recommend you use data that is 90 days or less old. Using older data than this may target an incorrect audience.
why are there percentages on a lookalike audience?
A percentage filter will show up when you’re creating your lookalike audience. It ranges anywhere between 1 to 10 percent, 1 being the more laser-targeted option and 10 being the broader option.
Adespresso did a pretty awesome test to identify what the best most profitable percentage is, they found out the below.
percentages and results:
From their results, we can see that the more targeted 1% lookalike audience has a much cheaper cost-per-lead than the 5 and 10 percent.
Although this information was tried and tested over a reasonable amount of data, you should take it with a pinch of salt and try each lookalike percentage for yourself.
Remember, testing is everything when it comes to a successful digital marketing campaign.
how do facebook lookalike audiences work?
Now that I’ve talked a little about how Facebook Lookalike Audiences work, hopefully you’re ready to dive in and try them out yourself!
Lookalike Audiences can be an extremely powerful way to leverage Facebook’s algorithm and data to increase your traffic and conversions.
If you’d like to learn more, sign up for our newsletter below and become an e-commerce genius!