A few short weeks ago, I wrote a post waxing lyrical about Facebook lookalike audiences and their numerous targeting opportunities. If you haven’t seen it, you should. Check it out here.
Today, I want to take on lookalike audiences from a slightly different angle. That angle is, controversially, excluding them from your retargeting strategy.
See, when you’ve got your winning ad sets up and running, it’s tempting to cast your net as wide as possible. The more people you advertise to, the more conversions, right?
Sort of. If you’re advertising to people with no interest in your product, you’re literally squandering money on being ignored. In fact, with Facebook advertising being the way it is, excluding people from seeing your ads is a guaranteed way to succeed.
👉 Let’s Quickly Recap
I expect most of you know what a Facebook lookalike audience is, but here’s a quick reminder.
Lookalike audiences are built up of people likely to be interested in your product or service because they already match your best existing customers.
Lookalikes can reap you a world of benefits, from higher click-through-rates (CTR) to better quality leads.
Facebook’s algorithm does a fantastic job of finding these users, and you can enjoy the results with barely any effort on your part. You can even control the size of the lookalike audience you want, from 1 – 10%.
If you opt for a 1% lookalike audience, Facebook will find users with 90% of the same characteristics as your current customers. If you go right up to 10%, Facebook will show your ads to 10% of the population in the countries you choose.
A 10% audience is more likely to give you “wildcard” leads. Facebook will show your ads to users who wouldn’t usually see them but have just enough matching characteristics to give them a shot.
There are obviously pros and cons to opening up your audience this wide, so you’ll need to carefully monitor your lookalikes over 48 hours to see the results.
One major con with a broad lookalike is the same problem I mentioned earlier: advertising to people who’ll flatly ignore your product until it goes away. Which brings me to…
👀 Exclusion Targeting: A Sneak Peak
When other blogs talk about “exclusion tactics”, they’re usually talking about custom audiences.
As a starting point, this completely makes sense.
Facebook faithfully starts showing your ads to those within your specified parameters. These are usually people who’ve seen your ads before and made that initial click – but didn’t convert.
This is a great way to get conversions because some people take longer to make a buying decision than others. Chances are, the second or third time they see an ad, their resistance and will-power will break entirely, and they’ll end up making the purchase.
On the other hand, some people are attracted to a product or service but possess an iron will. No amount of retargeting will convince them to take out their wallet. If you continue to bash away at them, you’re virtually flushing money down the toilet.
It’s time to exclude these people and forget about them forever.
You don’t need these kinds of people in your life. Instead of wasting your budget on users who don’t want what you’re offering, use your Facebook pixel data to hone your audience into those who do. This is called “excluding a custom audience”.
Remember this, as it’ll help you with your lookalike audiences later on.
🐛 Before Excluding Prospects, Segment Them
When it comes to Facebook ads, some of the leads you get are going to be better than others. Properly segmenting them is the name of the game here. Done well, segmenting can have a massive effect on your overall lead quality.
I’ve spoken before about LeadsHook and what a successful investment it’s been for us.
We use the quiz software LeadsHook in most of our ad funnels. Though quizzes mean the audience will need to make more clicks, the reams of data help qualify our leads and categorise them according to type.
Say, for example, that you want leads for insurance or a loan. The user will click the ad, read the advertorial, and take the quiz on the landing page.
Depending on their answers, you can send these leads to two separate thank you pages. For the sake of ease, let’s call them Prospect A, Prospect B, “Good Thank You Page” and “Bad Thank You Page”.
Prospect A has a steady job, and they’re a homeowner. They’re looking for a large loan, and their credit rating is squeaky clean. It’s likely they’ll become a converted customer, so you send them to the “Good Thank You Page”. This thank you page will probably contain information about a scheduled call or email and will welcome them on board.
Prospect B, on the other hand, is a bit down on their luck. They’re currently renting, and they’re unemployed to boot. On top of that, they only want a small loan, and their credit report hardly sings their praises. Naturally, the quiz directs them to the “Bad Thank You Page”. This one will say something like “Sorry, we can’t help you this time,” and may offer some alternatives.
When it comes to Prospect B, there is a way to get less of these cluttering up your distribution system: excluding them entirely.
[But if you do get Prospect B-type leads, don’t mark them as useless just yet. There are ways to make money from Prospect Bs still, and you can find them here].
❌ And The Nitty-Gritty: Excluding Lookalike Audiences
Imagine you’ve got two people doing freelance work for your business. One is a delight; they’re creative, professional, and courteous.
The other person, on the other hand, is a disaster. Their work is shoddy, they consistently miss deadlines and, when you call them out on it, they’re obnoxious.
You know all this, and yet you continue to hire both of them anyway. Then you sit back and wonder why half your results are terrible.
This can happen if you don’t exclude someone from the outset.
So you want more Prospect As and no more Prospect Bs. The answer? Use your segmentation strategy to find two types of lookalike audience.
If you’re doing things right, you should see your current audience divided cleanly into two. Use your pixel information and create a custom audience of those who are hitting the “Good Thank You Page” and another of those reaching the “Bad”.
Once your pixels have enough data (about 300 hits on each page), you can then use this info to create a “Good Lookalike Audience” and “Bad Lookalike Audience”.
Next, go onto Ads Manager, select the campaigns you want to exclude from your “Bad Lookalike Audience” and click onto “ad set”. There, you’ll see a small “exclude” button. Click onto it and make sure you save your changes!
There you have it – anyone included in your “Bad Lookalike Audience” will not see ads from that campaign. Excluding is an excellent strategy, as you could potentially save yourself millions of wasted clicks. Best of all, Facebook is doing all of the hard work for you.
Now you’re building lookalike audiences on purely top-quality leads, you’ll see your CPL drop and could even charge your client more for your services.
So, there you have it – another Facebook ad hack from Flexxable. If you’re interested in more ways to reduce your CPL, scale your ad campaigns, and maximise your profits, head on over to www.flexxable.com. Tell us exactly what you’re looking for, and we’ll go all-out to help.
Got any more tips and tricks for lookalike audiences? Share them in the comments below!