New information on how various factors affect YouTube’s video recommendation algorithm will be released by members of the team who will be responsible for working on it.
After we were only implemented in 2016, we still have an overview of how YouTube’s machine learning algorithm works.
We know that video recommendations are influenced by factors such as clicks, time, likes / dislikes, comments, timeliness, and upload frequency.
For example, we don’t know if external traffic has an impact on video recommendations.
It is also unknown whether poor performance affects the likelihood of future videos being recommended.
The effects of other potentially negative factors such as inactive subscribers or too frequent uploads are also unknown.
These are the factors that the YouTube team will discuss in a new Q&A video on the recommendation algorithm. Here you will find a summary of all questions and answers.
Read on below
Badly performing videos
If one of my videos underperforms, will it harm my channel? Could a few bad videos get better videos in the future?
YouTube doesn’t rate any channel as a whole based on the performance of some videos.
YouTube only cares about how users react to a particular video when they decide whether to recommend it to others.
The recommendation algorithm will always “follow the audience”.
When a video attracts an audience, it will appear in users’ recommendations regardless of how the channel’s previous videos performed.
Channels shouldn’t worry about an algorithmic downgrade based on a drop in viewership.
It is normal for videos to vary in performance in terms of views and other metrics. Therefore, YouTube makes sure that not all recommendations depend on these metrics.
Read on below
Too many uploads per day
Is there a point where the number of videos per day / week on each channel is so high that the algorithm is overwhelmed and videos slip through?
There is no limit to how many videos a channel can recommend to a particular viewer in a single day.
Channels can upload as much as they want. The number of views each video receives depends on viewer preferences.
YouTube’s recommendation system will continue to recommend videos as long as viewers continue to watch them.
If a channel is uploading more videos than usual and each video is getting fewer and fewer views, it can be a sign that the audience is burned out.
While there is no limit to the number of videos YouTube recommends from a channel in a day, there is a limit to the number of notifications that are sent.
YouTube only allows 3 notifications per channel within 24 hours.
My channel has been around for a few years and I think I have a lot of inactive subscribers. Should I create a new channel and then re-upload the videos to be more acceptable to the algorithm?
Inactive subscribers have no control over YouTube’s recommendation algorithm.
This goes back to the first question that YouTube said its algorithm was following the audience.
A channel with inactive subscribers can still see their next video in the recommendation section if it attracts an audience.
Creating a new channel and re-uploading the same videos won’t help more people view those videos.
YouTube stores viewer preferences so there is little chance of reaching these inactive subscribers with a new channel.
Read on below
Creators should only start a new channel if they decide to go in a different direction with their content.
How important is external traffic?
External traffic is definitely a factor influencing YouTube’s recommendation algorithm.
However, their influence only extends so far.
External traffic can help cause a video to appear in the Recommendations section. But once it’s there, it has to do well with the audience.
The long-term success of a video depends on how users react after clicking on it in their recommendations.
I’m getting a lot of traffic from external websites, which is causing my click-through rates and average view times to drop. Will this affect the performance of my video?
According to YouTube, if a video receives a significant amount of external traffic, there is no problem if the average view time drops.
Read on below
Apparently, this is often the case and does not affect the long-term success of a video.
Again, the YouTube algorithm is more concerned with ensuring that viewers engage with a video after they click on it in their recommendations.
The algorithm does not deal with the actions of the viewer after they click on a video from an external website or app.
For more information on these points, see the full video below: