Back to the top
E-commerce Marketplaces Uncategorized

Are you shutting down potential sales on Facebook?

Find out if you are missing out on potential sales on Facebook, and discover tips to optimise your presence to increase conversions.

27 Jan 20254min. reading timeThomas HaurumThomas Haurum

If you are using lookalike audiences on Facebook, you could be limiting your potential sales on the platform.

Working with lookalike audiences on Facebook also means you are narrowing down your target group. In fact, Facebook is often better than you or I at targeting ads according to both campaign type and the objectives of your Facebook campaigns.

Here you will find practical tips on how to scale your sales on Facebook with and without using lookalike audiences, as well as how to test your lookalike audience against a cold audience.

What are lookalike audiences on Facebook?

Lookalike audiences are groups on Facebook that resemble people who are ALREADY in your existing audience. Lookalike audiences allow you to narrow your reach in terms of the number of people to whom you show your ads.

You can create lookalike audiences on Facebook with sizes ranging from 1-10%, which are similar to your current audiences.

Your current audiences could be, for example:

  • Visitors to your website
  • All purchasers
  • Users who have added items to their basket
  • Anyone who has engaged on Facebook or Instagram
  • All newsletter subscribers
  • And so on…

For example, if you sell women’s shoes, you can create a lookalike audience based on previous buyers from your webshop.

You set up a lookalike audience in Facebook under Audiences > tick the audience you want to copy > choose audience location (country) > select audience size

  • 0-1% lookalike audience are those most similar to your target audience
  • 1-2% lookalike audience are those next most similar to your target audience
  • 2-3% lookalike audience are those somewhat less similar to your target audience
    and so on…

For example, you can create a 1-10% lookalike audience, which gives you a group of up to 433,200 people in Denmark.

Below is an example of a 0-1% Facebook lookalike audience, which is similar to all visitors in the last 60 days.

BUT with Facebook lookalike audiences, you risk missing out on potential sales! (OUCH OUCH)

BECAUSE! If you only advertise to a 1% lookalike audience, you are NOT reaching all the other potential buyers in your chosen country – and there are many. They could potentially be in the remaining 99%.

BECAUSE! Facebook’s algorithm might not function optimally in the learning phase (read about Facebook’s learning phase here)

AND! Your audience can become depleted (because it’s too small) + the frequency of ads shown to each person increases (making each sale more expensive).

AND! You don’t know if a 1% lookalike audience converts better than a 2% lookalike audience.

AND! You don’t know if a “cold” audience performs better than lookalike audiences.

Okay! So what should I do?

You should try allowing Facebook to manage ad delivery, so you focus LESS on the audience and let the AD CONTENT determine your SUCCESS.

In other words, you now test your ad content against different audiences, instead of the other way around (which has traditionally been the case). The previous approach (and still the case for some) was to choose demographic + interests for your audience and create a specific ad for them. This is very time-consuming (read: expensive). 

BECAUSE: Facebook knows BETTER than both YOU and ME who should see your ads (and with which message), based on the campaign objective you have selected 👇

Example of selecting a Facebook campaign objective

Isn’t there always a “but…”?

YES, there certainly is – as with almost everything else in online marketing. It might work, it might not. But if you don’t test, you might miss out on leads or sales, as you’re limiting both yourself (and Facebook) with lookalike audiences.

SO!
You need to work MUCH more with your ad content, try to focus less on audiences, and let Facebook do the work. Facebook’s algorithm is very good at delivering ads to the right people.

AND!
You should try split testing lookalike audiences against a “cold” audience.

How to use a cold audience for a Facebook campaign

BONUS (important): 
Facebook knows your previous visitors and buyers, and these will automatically be part of a “blank” audience. Facebook will therefore attempt to show ads to them first, as they are closer to making a purchase, meaning a “blank” audience will not function as intended. So remember to exclude, for example, previous visitors and your newsletter subscribers.
Example 1: A cold audience on Facebook, where no audiences are included

Using lookalike audiences for a Facebook campaign

In the same way as with a cold audience, you can include your lookalike audiences, which you want to target your Facebook ads towards.
BONUS (you read this earlier as well): 

Facebook knows your previous visitors and buyers, and these will automatically be part of a “blank” audience. Facebook will therefore attempt to show ads to them first, as they are closer to making a purchase, meaning a “blank” audience will not function as intended. So remember to exclude, for example, previous visitors and your newsletter subscribers.

Example 2: including lookalike audiences as targeting

Building a Facebook campaign with cold audiences and lookalike audiences

I would often recommend that you test lookalike audiences and a cold audience against each other within the same Facebook campaign, and let Facebook allocate the budget to the ad set which performs best. Remember to select “campaign budget optimisation” in your campaign settings.
Example of building 2 Facebook ad sets with a 10% lookalike audience and a cold audience.
The images above show 2 different ad sets, which could contain exactly the same ads. In the “lookalike audience 1-10%” ad set, you have the audiences from example 2 (see image), and in the “cold audience” ad set, you have the audiences from example 1 (see image).

“Stacking” Facebook audiences

You can also “stack” your lookalike audiences. Here you use one ad set for, for example, 0-1%, 1-2%, 3-4% and so on. You will notice that different lookalike audiences perform differently. “Stacking” is useful if you have smaller Facebook budgets to work with. Below shows an example of how you “stack” lookalike audiences, and how in this case the individual lookalike audiences perform. Here you can see that the 7-8% lookalike audience performs better than the 0-1% lookalike audience.
This was an overview of how you can work with Facebook lookalike audiences, cold audiences, and how you can “stack” your lookalike audiences to see what works best for your Facebook campaign. But REMEMBER to test and see what works for your specific Facebook campaigns.
Do you need help setting up, testing, and finding your best-performing Facebook campaigns? Feel free to contact us!

Do you want to see what your competitors and your industry are doing best?

At WeMarket, we offer businesses a benchmark report that compares their marketing efforts with their key competitors. You decide which competitors we should compare against.

We specialise in selling physical goods online and growing webshops – and now you can benefit from this expertise, even if you’re not already a client.

It’s completely free.

  • 1.

    Competitor screening

  • 2.

    Industry potential

  • 3.

    Recommendations