What is A/B Testing?
If you’re working in Digital Marketing, you must have already heard of A/B testing (or split testing). But let’s quickly recap what A/B testing means.
An A/B test is a process of comparing two versions of… well, anything really: Icons, Meta text, Screenshots, Web Page, Email, etc., you get the idea, to compare their performance.
In practice, this means one group of people call it Group A will receive one version of the asset you are testing, and another group call it Group B will receive the variation.
This process helps test the hypothesis and determine which asset works best – aka supports your goals better. In the case of a Youtube channel for example, it will help you determine which asset drives more views to your YouTube Channel or improve your audience engagement.
Before I go any further, if you’re new to Youtube Channel management, I invite you to read the Tips on starting a Youtube Channel – Becoming a Channel Manager.
How to perform an A/B test on Youtube?
If you already played around with the Youtube Dashboard, you may have noticed there isn’t a tool within the platform to run A/B tests.
So how do we go from there precisely?
You're willing to invest a little in a tool...
If you’re willing to put a little on the table, I find Tubebuddy to be an excellent tool.
You will be able to test your Titles, Thumbnails, Tags and Descriptions, so you can identify how to optimise your channel’s growth.
For example, on the same video, you will be able to test whether having text on your thumbnail generates a higher click-through rate, thus more traffic to your video, or else whether your video’s title affects your audience retention, etc.
You don't have any budget? ...it's time to get your hands dirty!
Investing in a tool is out of the question?
Well, there are ways around paying for a tool. There’s manual work you can do to perform A/B tests. One disclaimer, though, results won’t be as accurate as with a tool.
What I typically do, if I cannot use a tool to run A/B tests, is run one version of my asset for 2 weeks and run the variation for the following 2 weeks. When using this method, we do have to be really careful about seasonal fluctuations.
Advice: If you’re testing theories to identify how to optimise your click-through rate on a video, make sure to compare what’s comparable. Suppose you have a 15% CTR on a video that generated 1000 views in two weeks and a 14% CTR on the same video that generated 100.000 views in the following two weeks. In that case, you might want to re-test your two variations to compare against a more similar traffic to draw more accurate conclusions.
Why should you A/B test on Youtube?
Over 2 billion logged-in users visit YouTube each month, and every day people watch over a billion hours of video and generate billions of views.
Easy to say Youtube is a highly competitive field and, if you’re just about putting your content out there and hoping for the best, no matter how great your content is, this might not be the right approach…
Today, you need to be able to continuously optimise how you market your content. You want to find the best combination of thumbnails, titles, descriptions and tags that helps you drive the most traffic and keep your audience engaged. That is how you will ensure organic growth over time.
