ManuMercurial
patreon
YT Insights #2 - The 3rd Hidden Signal: Viewer Satisfaction (this is real)
🕑 Added 2023-08-14 16:00:04 +0000 UTCThere is a third hidden signal that the algorithm uses to detect a promising video, we’ll look at its history, and why we can’t really measure it.
Welcome to ‘Viewer Satisfaction’.
Very quick history lesson to understand why this even exists:
In the early days of YouTube there was no algorithm, everything was based on the number of views a video had. People quickly figured out that the more click-baity you could be, the more likely to get recommended.
There was no downside to click baiting viewers as more clicks on your video meant it would be shown to more people. This led to a lot of problems, as people were being tricked into watching videos that they didn't want to watch.
Likes did not help either as they could be easily exploited both for good and evil.
Then with the launch of the Recommendation Algorithm, we switched from a Views based system, to an algorithm focused on creating the longest Watch Session possible.
Enter the era of Youtube Rabbit Holes.
Youtubers quickly realize that the longer they can keep you watching, the more views they’ll get on their videos. I’m sure you remember this. The algorithm was so fine-tuned that every time you went into YT you would end up on hour long sessions going from one video to another, deeper and deeper into topics you didn’t even wanted to learn about.
But you always ended up feeling icky afterwards, like it was a giant waste of time.
That’s the reason they introduced ‘Viewer Satisfaction’.
We might be able to trick people into clicking on a video with clickbait, and then trick them to watch longer with other retention tricks, but if the video makes them dissatisfied, the algorithm will Not share it.
And how do they measure it?
Through Surveys!
Believe it or not, these surveys are one of the important factors for a video’s success today.
The algorithm is always evolving, all we can do to control the ratings we get is make the best videos we possibly can.
I know this is easier said than done (and we'll explore ways to achieve this). But one key take away is right now is the best time to focus on Quality over Quantity, as the algorithm favors videos that have a higher impact on our viewers.
Hopefully this history lesson also explains why CTR and AVD are still key metrics in a video’s performance!
Next week we’ll talk about how the Algorithm works.
Talk soon!
