Are you split testing your Facebook ads?
If not, I want this blog to help you understand why you should split test almost every variable in your campaign… which can ultimately help you save money in the future.
To understand how split testing works on Facebook, you’ve first got to understand the basic structure of a Facebook Campaign.
There are 3 layers that make up a Facebook Ad Campaign; Campaigns, Ad Sets & Ads…
Layer 1 – Campaign
This is where you set your objective and you can only have one objective per campaign. Your campaign objective should align with your marketing objective. You’re essentially telling Facebook what your goal is, and Facebook will then deliver your Ads to people most likely to fulfill that goal.
Layer 2 – Ad Set
In each Campaign, you have at least 1 Ad Set. This is where you set your budget, placement, and targeting. You can have multiple Ad Sets within a Campaign to allow you to split test your placement and targeting.
Layer 3 – Ad
This is where you create your Ad, upload your image, write text and add a link to your website or landing page. You can have multiple Ads within an Ad Set to allow you to split test your Ad Creative.
The layers allow you to split test!
Split testing means testing multiple variables in a campaign to determine which variable performs the best.
SPLIT TESTING ADS
Create multiple Ads in one Ad Set and split test your creative, including:
- Format (Video, Carousel, Single Image, Collection)
- Landing Pages
SPLIT TESTING AD SETS
Create multiple Ad Sets within your Campaigns and split test:
- Placement (Desktop, mobile, Facebook, Instagram, instant articles, right-hand column, messenger)
- Targeting (Location, age, gender, demographics, interests, behaviours, website traffic, engagement, email lists, lookalikes)
- Optimisation (Optimising campaigns for engagement, link clicks, landing page views or custom conversions)
I know what you’re thinking… “THIS SOUNDS HARD – why should I bother.”
…Because you’ll save money!
You’ll start to see, first hand, what WORKS and what DOESN’T WORK…
This will allow you to spend money on the audiences, platforms, and ads that give you the best return on investment.
Here are a few rules when it comes to split testing for beginners…
Start with one platform
If you’re starting with a small budget, start with mastering one platform, whether that be Facebook, Instagram, Suggested videos or Messenger.
The platform will determine the strategy for your ad, along with the creative and copy.
Split test at least 2 audiences
I always recommend starting with split testing 2 audiences… because targeting is pretty much the SINGLE MOST IMPORTANT component of any campaign.
Each Ad Set requires a minimum budget, which is why I’ve suggested only split testing 2 audiences… If you have a larger budget, split test more.
If you find one audience is not performing well, turn that Ad Set off and create a new Ad Set testing a new audience.
Split test your creative EVERY TIME!
It doesn’t cost any extra to have up to 6 different styles of Ads delivered to your audience. This also means that your audience will see a variety of your Ads in their newsfeed… which helps keep things interesting!
Start by split testing your format to see which performs the best, then once you’ve found a format that converts well, start making tweaks to the images and copy until you find yourself a winner!
Ads won’t be treated equally…
When split testing Ads – Facebook will determine within just a few hours which Ad is performing the best and will then distribute most of your budget to a single Ad. This means the split testing creative from the Ad Layer is not always accurate. If you want an accurate split test of your creative, create separate Ad Sets with just one Ad for each Ad Set.
If you’re just beginning with a small budget though, there’s no harm in grouping Ads together.
That’s it folks.
I hope you enjoyed this introduction to split testing, If you have any questions, feel free to shout!