- The Real World Marketer
- Posts
- MrBeast Shows Us How to A/B Test the Right Way
MrBeast Shows Us How to A/B Test the Right Way
Do you know how to a/b test like a master?
MrBeast Shows Us How to A/B Test the Right Way
MrBeast is known to most people who use the internet or YouTube. But what many people don't realize is that he's also an incredibly intelligent online marketer. He recently performed an A/B test that proved my hypothesis.
Check out this recent thumbnail test he did.

Jimmy recently decided to run an A/B test to see if the image he placed on his thumbnails had an impact. He went very granular, using the exact same images except for his facial expression. He had a hunch that having a smile instead of a mouth hanging wide open would get more views and action.
The only way to effectively test this without skewing the results was to keep every other variable the same. He did exactly that. But what if it was a fluke? (Irony considering that this saying is to advocate for something uncommon when a "fluke" is one of the most common fish in the sea.)
Jimmy decided to test it again on another video, again keeping everything the same but his facial expression. And what did it prove?
Captain Jimmy’s smile is what the people want. 🙂

Another important thing to note is that Jimmy kept his outfit the same. Despite the background change, he looks relatively the same as in his previous A/B test.
It's important to set up your A/B test so that you're testing one element at a time. If you change more than one variable, you could skew the results. Jimmy is clearly a master of A/B testing, as evidenced by his success on YouTube. This one result proves my hypothesis that he's not just a creator, but a literal marketing genius.
I often see people running A/B tests with more than one variable at a time. Please don't be that person! If you want to see if two entirely different landing pages are better than your current landing page, that's fine. However, if you want to see if a blue CTA gets more clicks than a red CTA, don't change the text in the CTA or on the page at all.
When you're running your next A/B test, remember this: keep it one piece at a time, like Johnny Cash.
At the end of the year, the subscriber with the most referrals will receive a special prize from me.
Reply