Walk Through How We Set up A/B Tests for Our Meta (Facebook) Ad

Twelfth in a series

After tweaking the graphics and studying the search term data, we were ready to create our Meta (Facebook) ad. Follow along to see why we made the choices we did.

First of all, we have to remind you that the only constant about Meta is that it’s constantly changing. We create ads on the platforms of different clients and the experience differs depending on whether the latest changes have rolled out for their account or not. We’ll give you the general gist, although your experience may be a little different.

For this experiment, we are running a total of four ads and comparing them. The first A/B test includes Graphic Version A with two versions of primary text. The second A/B test includes the same two versions of primary text, but with Graphic Version B. The goal is to determine which graphic and text combination generates the most interest.  

Choosing your objective

The start of the ad creation process asks you to choose your objective. Depending on whether Meta has updated your page yet, the choices will be different. Meta says “We are gradually introducing a new set of 6 simplified campaign objectives in Meta Ads Manager. While some objective names and campaign creation steps are changing, all the same functionality will be available” and they have a web page to further explain things.

Do read through the objectives carefully because what sounds like the perfect objective may not mean what you think it does. For instance, while we are looking for “leads,” we chose “traffic” as our goal instead because we want folks to come to the landing page for the SEO audit. We can better demonstrate the advantages of the audit and our trustworthiness there than in a single ad.

Choosing your ad type

As we’ve said earlier, we decided to try a Slideshow Ad because it has movement like a video but is simpler to create than a video. It’s not immediately apparent how to choose “Slideshow,” but you’ll find it after you click on “Single ad or video.”

There are many fun things you can do when creating a Slideshow Ad and we suggest you read Meta’s Help Center page about it. If you want, you can create your ad from photos or images you posted on your page in the past or you can upload new graphics. You can crop your images and add things such as your business logo or a text overlay.

Once you arrange the images to follow your marketing story, decide how many seconds you want to spend on each “scene” and what you want your scene transitions to look like. Each “scene” will have the same number of seconds and the ad itself can be no more than fifteen seconds long. Transitions choices are “None” or “Fade.”

Choosing A/B tests

There’s a button for A/B tests early in the process. Click it if you want to run an A/B test, but don’t worry if you don’t see any details for a long time. You’ll run through the entire process for version A and publish it before you get the option to work on version B.

Choosing a budget

Since ad placement is all about algorithms these days, you need to generate enough data for the algorithms to do their thing. Plan on running your ad for a couple of weeks and spending around $10 a day to get a big enough sample to analyze the efficacy of your ad.  

Choosing a Call To Action

This is another place where Meta has been making changes. There are now a whole bunch of CTA buttons to choose from. We suggest you go through the list to find the one that works best for your objective. We’re trying out “Get Offer” for this SEO audit, but we might make another A/B test with a different CTA in the future.

Choosing a Headline, Primary Text, and Description

These are additions to the graphics. For this part, the keyword research we did when creating the landing page for our SEO audit offer came into play once again. Since you only have limited space, make sure these are the best words for engaging your viewer.

The Primary Text appears above your graphic. It looks like what you’d write in a Facebook post. The Headline appears below the graphic, along with the Description. These look like the snippets you see when you share an article on Facebook. The Call To Action you chose previously will appear as a button across from the Headline.

Keep in mind that people are not on Facebook to look at your ad. Since you are trying to divert their attention, keep everything eye-catching and quickly scannable. Depending on what device your viewer is using, text may be truncated, so keep it short and put the important words up front.

When you have run through all the options, you will be asked to publish your ad. You can choose to make it live immediately or schedule it for another time. Once published, you’ll get the option to do it all over again for your B version to test against the A version. Meta says “Test will run 4 days unless early winner is found,” so be sure to check back.

There are more options to discuss, but this is plenty for now. If you’re feeling overwhelmed, you don’t have to do this alone. Our team can do the heavy lifting for you. Give us a call today!

Kate GingoldKate Gingold

I have been writing a blog with web marketing tips and techniques every other week since 2003. In addition to blogging and client content writing, I write books and a blog on local history.

Other posts by Kate Gingold

Contact author Full biography

Full biography

I have been writing a blog with web marketing tips and techniques every other week since 2003. In addition to blogging and client content writing, I write books and a blog on local history.

x

Linked In Google Plus