A/B Testing for Video: Measuring Results in Facebook

Megan O'Neill


In the last installment of out A/B Testing for Video series, we showed you how to set up a video A/B test in Facebook. Today, we’ll show you how to understand the results of your test. Which video performed better? Read on to learn how you can figure out.

A/B Testing Video Results

Reading results in Facebook Ad Manager

Luckily, determining the winner of your A/B test is super easy if you’ve followed the instructions for setting up your test that we provided in the previous post. Simply go back into the Ads Manager and look for the “Campaign” you set up. Click on it to open up your Ad Sets in your Campaign. There should be 2 Ad Sets listed – one for each video.

In the screenshot below, we’re viewing the results from a test in which video variation A began with a video clip and video variation B began with a photo. As you can see, the video that had a video clip at the start received 7X as many link clicks and nearly 6X the reach. The video clip start – variation A – is the clear winner.

Facebook Ad Set


If, when you get into the Ads Manager and click to open your campaign and aren’t seeing any results next to your ad set, or aren’t seeing the information you’re looking for, here are a couple of things you can try:

Note seeing any results? Check your date range.

Results will show based on a date range that you’ve selected. If this is the case, you’ll likely see the following message. Click on Change Date to set your date range. Make sure you select a range that begins before your campaign started and ends after it ended, in order to see the full results.

No activity during date range

If you do see results but want to change the date range to view results from only a select portion of the time your campaign was running, click on the dropdown menu in the upper right-hand side of the Ads Manager, under the Create Ad button. From there, you can select a specific date range or you can view results from the lifetime of the campaign, today, yesterday, this month, last month, or the last 7, 14, or 30 days.

Set date range

Not seeing the stat you’re looking for? Check your columns.

By default, Ads Manager will show you performance results first. This, as shown in the first screenshot, includes the metric you optimized for when you set up your ad (in this particular case, link clicks), as well cost per the metric you optimized for, and reach.

If you are only testing to see which video performs best for a single objective (like link clicks), this should be enough. However, if you’re looking to see how your videos did in terms of other objectives (perhaps likes or shares), then you’ll have to add a different set of columns. You can do this by opening up the dropdown on the right that says “Columns: Performance.”

Performance Columns Facebook Ads Manager

Once the dropdown menu is open, you can either select a different pre-set from the list (like Engagement, Video Engagement, or Performance and Clicks), or you can “Customize Columns” and then check off all of the different columns you’d like to add to your view.

Customize Columns Facebook Ads

What’s next?

Once you’ve determined which video variation is the winner, you may be wondering what to do next. Well, you’ve got a couple options. You can either move forward with your winner or continue testing.

Option 1: Move forward with your winner

If you’re happy with your winning video, congratulations! Stop the spend on the losing variation and put your ad spend into the video that’s giving you the biggest bang for your buck. You can stop an ad variation from running by toggling the switch to the left of the variation in the Ads Manager to the off position.

Option 2: A/B test your winner against another variation

You can also go back to the drawing board to see if you can make your video even better. If you need ideas, check out our post on what variations to A/B test.

Whatever you decide to do, it’s important to keep track of all your tests, remember which variations performed best, and incorporate them into your video marketing strategy going forward. For instance, the test features in this blog post showed that starting with a video clip outperformed starting with an image. If you ran a test with similar results, you may decide to do a similar test with your next video – A/B testing a version starting with a photo and one starting with a video clip. If you saw similar results, starting out your videos with video clips may be something that you start doing with all of your videos going forward.

Have you tried A/B testing a video on Facebook? If so, we’d love to hear how it went! Leave a comment here or reach out to us on Facebook or Twitter. And stay tuned for the next post in our A/B testing series, where we dive into A/B testing TrueView ads on YouTube.

Explore the other posts in this series:

What is A/B Testing? What Variations to Test
Creating Video Variations with Animoto Setting Up a Test in Facebook