As your campaigns progress, you’ll most likely ask yourself:
What prompted the user to click? Was it the text, image, or video? I'd love to know!
I'm going to show you how you can know using the A/B testing!
Make the Most of A/B Testing, a Powerful Tool
It’s impossible to know all the factors that prompted a person to click on your ad simply because you’re not actually with them, in the same room, looking over their shoulder. 🙂 Still, you can create two very similar ads, only changing one parameter between them, to test each parameter individually.
Does this approach vaguely remind you of science class? Not surprising because A/B testing is based on best practices of empirical studies: varying a single parameter in two otherwise identical contexts to fine-tune the results and draw the most accurate conclusions possible.
For example, if I can’t decide between two different messages, I can create two ads that are identical in every way except for the message. This way, I can accurately measure which variant is the most successful.
This is called A/B testing because you test one variant (A) against another (B).
Avoid creating too many combinations for one campaign. Just ask yourself what parameters you want to test. I recommend working on the message and the image associated with it for your first variations.
Let's take another example. You want to promote an online course for Facebook ads, and you have come up with these two slogans: "Boost your Facebook ads performance with our online course" or "Facebook ads not performing as you’d like? See how to get better results with our online course."
You also found these two photographs:
Does that mean I have to create four ads from scratch?!
Well, yes, unfortunately, the Facebook interface is not very well-suited for A/B testing. I suggest you use AdEspresso (which I have already mentioned briefly) once you really want to do A/B testing. This tool will save you hours!
Create Your Campaign on AdEspresso
AdEspresso is a tool for managing your Facebook campaigns in a much more organized, efficient, and fun way. It's not free, but it does offer a 14-day free trial period, enough time to test out your new campaign!
Click on Signup to create a free account.
To create your first A/B testing campaign, click on New Campaign in your dashboard. The ad creation page will then be displayed.
Give your campaign a name, then select the objective based on what you want to promote: a website, Page, or mobile app. Then click on Proceed.
Configure your ad by selecting the format, Page, title, associated message, images, link, and anything else you'd like to add.
Create your first ad and preview it at the bottom of the page.
When you’re satisfied with what you’ve made, return to the top of the page to configure your variations.
You will see several blue crosses on the right side of most of the fields on this page. They show the parameters you can use for your variations.
In our case, we want to test both the message and the associated visual. So we would click on the plus sign to the right of Ad text and add our variation. As for the images, we can add one by clicking on Upload new images.
If you scroll down the page and click on the blue arrows, you will see the new variants!
Select the locations by clicking the blue check mark to the left: Desktop Feed, Mobile Feed, Network Feed, Instagram, and Right Column.
Ultimately, these additional parameters give you more options than the Facebook tool.
AdEspresso offers you a simple way to track your campaigns with Google Analytics (if you use it). You can leave the default values.
If you want to set objectives that use the Pixel, check the last box and select your Pixel.
Then click on Proceed.
In the next window, set up your audience the same way you would on Facebook and click Proceed.
AdEspresso also lets you test your audience, i.e., determine which audience reacts best to your ad. I don't recommend testing both your audience and ads. The best way to refine your audience is directly via Facebook (which we went over in the previous chapter), as it produces better quality and is tailored to your business.
So here, select the placement and then click Proceed.
Just like on Facebook, decide on your budget and click on Proceed, then Publish.
Your campaign has launched! Well done!
Analyze Your Results
Click on your campaign, then All Ads, to see any preliminary results.
You’ll undoubtedly notice that one of your ads was more successful than the other(s). You should then optimize it the same way you would an ad on the Facebook ad management platform: adjust the budget and reply to comments as soon as possible.
Stop A/B Testing at the Right Time
You will notice very quickly that differences in performance are growing between your ads, and you may be tempted to let them run until the end of the campaign.
But by doing so, you can quash the engagement a user may have with your campaign. The more settings you test, the more different ads a user will see from the same campaign! While they will certainly interact with the first one, they are unlikely to do so a second time.
This means you need to spot the ads that are working well as soon as possible and stop the ones that aren't converting. This will help promote the one that works and increase its conversion rate.
To assess the conversion percentages between two ads, I suggest using the A/B Significance Test, a tool developed by Neil Patel.
When your experimentation is complete, keep a chart of the factors you tested and the best practices you noticed. This will help you for the future!
Let's Recap!
A good way to optimize your campaigns and get better results is to make use of A/B testing, which allows you to:
Test various elements (text, visual, landing page, etc.).
Identify what works best.
Build more effective ads based on your test results.
You are now ready to optimize your campaigns and get the best possible results on Facebook and related networks. Ready, set, go!
