Add Structure To Your Facebook Ad Split Testing

Working at Hello Social means that you get a pretty intimate understanding of Facebook advertising.  We run a lot of campaigns for a variety of massive and minuscule brands, but for all of our clients split testing (aka. A/B testing) is an invariable part of our strategy.

In all of our various relationships, it’s pretty common that brands will understand that their campaigns should be “optimised”, but might have very little idea on how that is practically achieved.

The real challenge in split testing on Facebook is creating meaningful tests that give you actionable results. While it is all well and good to run some different ad creatives against each other, this is not a structured or strategic approach to split testing.

We’ve created this awesome free report template that will save you a bunch of time spent wondering how best to record your test results. Using the Hello Social Facebook Split Testing Spreadsheet you can easily track, filter and analyse your results, referring back to past tests so that you keep on top of the data.

If you are a Facebook A/B Testing newbie, or a more seasoned marketing manager, we hope that our tool will give you a kick start with the strategy and structure you need to impress your clients and team by being a nerdy know-it-all.

To explain exactly how you can start using this tool today, watch our 5 minute video tutorial on how it works.

The video tutorial covers a pretty solid technique for recording, organising and planning your Facebook ad split tests, but all of your optimisation decisions will require an understanding of some important metrics.  We’ve explained some of them below.

A good split test is reliant on being measurable.  While “conversions” are of course central to your campaign performance, they are not the be all and end of all of social media measurement, particularly when it comes to split testing.

Here’s a brief round up of some of the important social media metrics that are included in our split testing spreadsheet.

Conversions – Pretty self explanatory, but important nonetheless. These are often the primary goals of your digital marketing campaigns, and could be anything from a product sale, to a mailing list subscription or another type of engagement.  Using our tool, your conversion goal should be linked to whatever your campaign objective is.
For example, if you are running traffic ads, your conversion might be a sale or a lead on your site.

If you are running the brand awareness objective, your conversion might be a page like.

Engagements – These are conversions that happen on the Facebook platform which are broadly tracked as engagements. These can be split down into their various forms, from page likes, to post engagements, shares and comments.
These might be the main focus of your campaigns, so don’t think that they are less important because we’re calling them engagements rather than conversions!

Cost per engagement – This is the cost per post/ad engagement that happens on the Facebook platform. Note that this is a cumulative metric that includes not only clicks on links, but also likes, shares, comments etc.

Cost per link click – Often confused for the more readily available engagement metric, this is the cost that you paid for clicks on links embedded in your post. This could be the main linked image or video on the post itself, the CTA button on some formats, or a link that has been included in the text description.

Cost per conversion – You guessed it, this is the amount spent divided by the number of conversions. This gives you your cost per acquisition/conversion/order and is always one of the most important metrics.

Reach – How many individual people were served your ad, regardless of engagement.

Impressions – How many times your ad was shown on a screen to users. This is a different metric to reach as it counts multiple views for a single user.

Click through rate (CTR) – This is the number of clicks on your ad, divided by the number of impressions.

Cost/Revenue Ratio – This is an alternative calculation to the widely used Return On Ad Spend (basically the reverse). This is simply the amount spent or total cost, divided by the revenue generated.  Lower ratios are better, and anything above 100% is a loss. If you prefer ROAS, then just flip the calculation around.

These certainly are not all the metrics that you could track, but are in my opinion the most important KPIs that you should be using to track your campaigns’ success.

The possibilities of split testing are almost endless, which is both a blessing and a curse. Testing is a huge part of social media marketing, and taking a structured approach makes all the difference, but be careful not to get caught up in creating too many variations as you will end up with a data set that takes too long to analyse.

Our top tip is to start simple, and focus on pulling one lever at a time. That way your
learnings will be easy to record and apply to your next test, and before you know it your campaigns will be giving you a better return than ever.