Will Better Testing Save Marketer's Organic Video Results?
Finally, the change we all wanted to see has started to land on our shores!
Nobody likes to deal with Facebook Creator Studio, but hopefully, this update will bring smiles to faces and some sunshine to the hellscape.
We all know that Facebook video engagement is down thanks to the great video cull of 2018. Before that, Facebook video was the most fertile of lands for marketers. Maybe inspired by the success of TikTok and Instagram's push into Reels, Facebook is now going to make it a lot easier to ramp up video views.
A/B testing will change the game
That's because Facebook has launched a tool that allows users to A/B test their videos before publishing the most successful variation.
Testing is something the community has been crying out for, and it could help breathe some life back into Facebook videos, which were sidelined for nearly three years.
The majority of people who have access to the Creator Studio will get their hands on the update now. It works by running two variations of the same organic video post until a winner (in terms of engagement metrics) is reached.
These test variants stay in the Creator Studio – the losing posts hang around the system but will not be published to your Page, and there'll always be a record of their performance in Creator Studio. So there's the chance to learn from the winning variation, but also the losing variation.
How will the testing work?
"Organic Video Post Testing in Creator Studio lets you test 2-4 variations of a non-ad video post to see which one your audience likes most. During the test, the posts are circulated to your audience, but not posted to your Page. When the test ends, the winning post is automatically published on your Page."
Once Facebook has identified the most successful post, it will automatically publish it – unless a user has scheduled the upload for a specific time.
Users will be able to test a variety of things:
- Emojis vs no emojis
- Long vs short descriptions
- Aspect ratios
- Different thumbnails
What is the process for use?
Each test will be shown to a subset of your Page audience, and admins will get to chose which metric they'll use as a determination of success. For example, you could pick one-minute views as the key metric – whichever variant gets the most one-minute views will be crowned the winner.
However, once the test is done, you will still see all the numbers for both variants. This should be super helpful as it'll allow marketers to learn more about which different creative decisions impact various aspects of distribution and engagement.
Here's how you can get the ball rolling on these split tests
- Click the blue Start Test button
- Select the Page you want to run the test on
- After you've selected the Page, go to the Create Post Test tray.
- In the tray, each row (A, B, D and C) represents a variation of your test.
To create an organic post-test, you'll need to:
- Click Upload in the first row and select your video file
- Click the New button to insert a new role, or Duplicate to copy one of your existing variants into a new slot.
- Once the video is uploaded, you can edit the title, description and thumbnail in each row to test these aspects of the video.
- Click the box under Default to choose a default winning posts – this is in case the margins are too small for Facebook to determine a victor.
Why this is exciting and how it'll help
Organic video testing will be a gamechanger for social media managers and marketers. It'll give you some serious firepower behind how you make decisions.
It'll help everyone learn what their audience wants and how to serve that. It'll also help the content get the reach it deserves because nothing is worse than spending hours on something you're really proud of to see it flop. The new feature should allow social media managers to improve page performance almost straight away.
But follow these best practices as well
Facebook recommends that you run a test for 30 minutes, which is the default. But you still need to work out what duration works for you. It's important to run the test for the shortest amount of time to get signals – a smaller page might need an hour, a bigger one just 10 minutes.
Don't run it too long because that exposes people to the losing variation of the test, so just run it long enough to get a signal of which variant will win.
You can run tests for 10 minutes, 30 minutes, one hour, three hours or 24 hours.
Facebook also advise that you run multiple tests before committing to a long-term creative strategy based on the outcome of those tests.
Whisper it quietly, but organic video results on Facebook MIGHT be finally back in business.