Facebook Ads data segmentation: how to use granular data to optimize your campaigns

Running a successful Facebook Ads campaign is no easy task.

Beginners often fall into the trap of thinking that because it’s easy to set up a campaign, it’s easy to get results. (Hint: that’s a costly lesson right there.)

Today, we’ll be focusing on a lesser-known optimization trick: data segmentation in Facebook Ads.

And it all comes down to testing. ABT all the way baby! (Always Be Testing.)

So…

Basically, we’re going into the Facebook data like a surgeon to find out what is working and what’s not — all so that we can double down on top-performing ads and cut the fat.

Long story short, you really need this stuff to be effective at Facebook Ads, so let’s jump right into it.

What is data segmentation — and why use it?

There are a few different definitions of this but they’re all relatively talking about the same thing. 

Data segmentation is the process of breaking up your data into segments so that you can analyze each one individually to look for insights.

These insights will tell you valuable info that you can use to optimize your Facebook Ads campaigns to get more leads, demos, free trials, purchases, and/or new customers.

For example, instead of just looking at how your campaign is performing on a high-level view, actually looking at specific segments to identify what image creatives, text copy, targeting, positioning, etc. is actually resonating with your target audience.

Top two reasons to focus on data segmentation

1. Trim the fat off your campaigns

It’s not just about finding out who engages with your ads the most, it’s also about who is engaging the least. Segmenting your data allows you to see which demographics aren’t giving you your money’s worth so that you can cut them out of your targeting strategy and spend your budget more efficiently.

2. Better campaign management

As boring as this benefit is, it’s important. The longer you run a Facebook Ads campaign, the more data you’ll get from your audience. Properly storing this data so that it’s easily accessible and easy to find out specifics is key, and segmenting that data is equally as important.

The bottom line is that using data segmentation allows you to figure out the exact type of person you should be targeting.

That way you can launch new campaigns and more data-backed experiments to continuously increase your ROAS (means “return on ad spend” for you newbies out there) to new highs!

With this, you can increase your leads, get more sales, and basically put more money in your pocket. 

Bam! Who doesn’t want more money? 

Let’s keep going.

Types of Facebook Ads segmentation

Facebook Ads can get fairly detailed when it comes to segmenting out your data, and that’s not referring to building custom reports (but let’s keep this actionable and not go on a technical nerdy rant).

The main types of segmentation we usually focus on for our clients are the following:

1. Demographic segmentation

Demographic segmentation is when you split your audience up based on factors such as their gender, age, income, or profession.

Maybe certain age groups are more likely to buy your product or if you’re selling a high-end product, higher-income brackets are better suited for your ads. Or, if you’re a B2B company, you’ll probably want to target certain professions. Facebook’s targeting allows for all of this.

2. Geographic segmentation

Geographic segmentation is something that every campaign should have, especially for smaller or local businesses.

The chances are that you won’t want to target users from all over the globe, so refining your targeting to only specific locations (e.g. countries, states, cities, or zip codes) is essential.

Hyper-specific geographic targeting does have some flaws, though.

If you get too specific, your audience size can become really small and this will hinder your campaign’s performance. Keep that in mind.

3. Device and placement segmentation

Self-explanatory but you can basically segment your data by what platforms/feeds/etc. your ads showed up on to get an idea of top-performing placements. Newsfeed is usually great.

4. Ad creatives and text copy segmentation

This one can drive you up a wall because there are so many things to test. 

The options are pretty much infinite. 

• Do you want to have one ad showcasing X feature of your product or Y feature?
• Do you want the color to be blue or yellow?
• What’s the CTA in the image going to be?
• Image or video?

This is really where you should be spending a lot of your time when it comes to advertising. Facebook’s algorithm has gotten so good that it can optimize your ad campaigns for you.

Meaning after you get 50 or so conversions, they can really start to know who to suggest your ads to. You can even try things like open targeting to get a cheap CPM after a certain point.

So one of the biggest levers you can pull these days is A/B testing your ad creatives.

Then segmenting out that data to see what experiments worked and which ones flat out sucked.

5. Detailed targeting segmentation

Interest-based targeting is older than my mom (love you, mom ❤️) so this is going to be some of the highest cost CPMs you’ll pay on the platform. So buckle up.

This type of targeting is super important if you’re just starting out because you don’t have any conversion data or other types of valuable data to even segment for.

At a certain point you want to start leveraging open targeting (if your data shows you it’s working, that’s why segmentation is important!) so that you can escape these high costs.

Now, these different segmentation options obviously aren’t every single type under the Facebook Ads platform — just the big ones.

(I don’t want to bore you with an exhaustive list of the data you track so let’s just focus on the ones you’ll need the most).

How to optimize your Facebook Ads with segmented data

Here’s where the fun part begins (if you’re a whackjob like me that loves data).

We’re going to be using a bit of logic and reasoning to find out what types of campaigns, targeting, ads, offers, etc. are performing well but most importantly – why?

By inferring from what the data shows, the why will help you build out more campaigns based on what is already working from your earlier experiments.

Supermetrics can really help out with the data analyzing portions of this because they directly integrate into the Facebook Ads platform.

We’ve even built customized reports that pull specific data from Facebook Ads in the past to better optimize our clients’ campaigns.

Segment your ad sets

Here’s how you can use segmentation to optimize your Facebook Ads campaigns. You’ll want to get really good at spotting trends and analyzing the campaign performance. 

Here’s an example of how to look at the data.

Segment your ad creatives

Creatives should get special treatment since they’re the most powerful weapon in your arsenal. I would spend a lot more time creating new variations and testing out different assumptions here.

One great ad creative can skyrocket your ad performance. Keep that in mind.

What can segmentation tell us about our Facebook Ads?

So, what exactly can segmentation tell us about our Facebook Ads?

The main things that it tells us are:

• Who the most profitable segments of our audience are
• Which types of ads those segments respond to

…and these might seem like fairly basic things to find out, but proper data segmentation takes this to a whole new level.

3 steps to improve your campaign with segmentation

Before we wrap things up we’ll leave you with some actionable tips that you can use to improve your campaigns with data segmentation.

1. Use Supermetrics for Google Sheets

Like we said, organizing your data is key and with Supermetrics for Google Sheets, you can easily export all of your Facebook Ads campaign data to Google Sheets. From there, you can format it as you see fit and actually use the data you’ve collected to your advantage.

2. Prioritize your segments

One approach you can take is to cut out the less profitable segments completely, and this is perfectly reasonable.

There are two other options though:

• Move less profitable segments into a separate campaign and try different types of ads to see what appeals more to them and work on turning them into a profitable audience.
• Reduce the percentage of your budget being spent on less profitable segments so that the more profitable segments are allocated a bigger portion of your budget.

For smaller businesses who don’t have a big budget to work with, it’s definitely worthwhile just eliminating any segments that are losing you money. On the other hand, if you have the money to work with them, with time you can turn these segments around and make them profitable.

3. Hyper-targeted ads

And finally, with all of this data that you’ve accumulated, you will be able to craft ads specifically for each segment of your audience.

This is where organizing your data with tools like Supermetrics for Google Sheets proves to be incredibly helpful.

With enough data, you can see exactly what appeals to the different groups of people in your audience and craft ads that specifically appeal to those groups.

All you have to do is create individual campaigns targeting specific segments, design your creatives, and start running your campaigns.

This goes hand-in-hand with budget management because for each segment, you’ll want to allocate a percentage of your budget in line with how profitable that segment has been.

Conclusion

There you have it — data segmentation in its entirety.

This is one aspect of marketing that amateurs don’t delve into simply because they don’t realize its importance, but it can make or break a campaign.

Using data segmentation will give you a huge advantage over your competition and if you take on board everything we’ve discussed above, including the strategies, you’ll undoubtedly blow them out of the water.

About the author

Lewis Mudrich is the founder and Chief “Idea Guy” at the B2B Ads Agency KonvertLab. They’re a B2B SaaS ads agency that grows companies using Google Ads, Facebook Ads, and LinkedIn Ads.