There was a time when internet users used to criticize YouTube’s algorithm for prioritizing views and clicks over ensuring that people get presented with quality content. Well, it looks like that time has finally passed and there are loads of facts (backed by YouTube itself) that confirm the same.
For starters, marketers used to believe that a single algorithm was responsible for ranking all the content there is on YouTube. This couldn’t be farther from the truth as the company has multiple algorithms in place for the said task. Different algorithms are used for ranking videos on the home page, suggested videos section, and the results page (with the former two relying on the recommendation system).
As per YouTube, its recommendation system is responsible for most traffic to hit the videos. This system is tasked with discovering videos for viewers, and not the other way around. In simpler words, the recommendation system fetches videos and sorts them in order of which ones a specific user is most likely to view. This, in turn, helps in boosting viewer satisfaction.
In fact, since 2015, viewer satisfaction has been perceived by YouTube as the top-ranking factor In 2012, watch time was the top-ranking factor and in 2011, clicks and views took the cake.
Ranking the videos as per their view and/or click count prompted creators to lure viewers in with the help of misleading titles and thumbnails. As for watch time, it was bumped down the list of the top-ranking factors after it was found out that this approach wasn’t good enough to tell if the viewers actually enjoyed the content they watched. Therefore, the decision to go with viewer satisfaction was taken.
What YouTube does is that it uses surveys and engagement signals to learn a user’s preferences. Engagement signals include metrics such as the likes, dislikes, clicks on the “not interested button,” average view duration, and average percentage viewed.
When it comes to the home page ranking algorithm, YouTube ensures that the most relevant suggestions are presented to each user. This is ensured via:
Another tip to retain viewers’ interest is by following a regular uploading schedule.
Now, let’s shed some light on the “suggested” section i.e. the list of videos on the screen’s right-hand portion while another one is playing. This section gives the user many options to choose their next clip to view.
Not only does the said section display videos that are similar to the one playing on the screen but also the ones relevant to the content the user was watching prior to that.
The suggested videos algorithm ranks content by taking the following signals into account:
Moreover, urging viewers to check out another video can prove to be helpful too. Last but not least, using playlists and end screens may increase the chances of your videos making it to the suggested section.
But here comes yet another twist i.e. external factors. One thing marketers should take note of is that external factors can affect their content’s rankings on both the home page as well as in the suggested section. And there’s barely anything that a creator can do about this hurdle.
Among the notable and uncontrollable external factors, the top ones are:
So, by accepting the fact that people’s interests change frequently, you may turn your attention to a different topic and see if that gets the numbers up.
Sources: How YouTube Works / Creator Insider - Behind the Algorithms / CI video 1 / CI video 2 / CI video 3.
Read next: Report Reveals Insights to Uplift Your Brand Reach and Audience On Instagram as The Marketing Strategy On the Social Network Changes in 2021
For starters, marketers used to believe that a single algorithm was responsible for ranking all the content there is on YouTube. This couldn’t be farther from the truth as the company has multiple algorithms in place for the said task. Different algorithms are used for ranking videos on the home page, suggested videos section, and the results page (with the former two relying on the recommendation system).
As per YouTube, its recommendation system is responsible for most traffic to hit the videos. This system is tasked with discovering videos for viewers, and not the other way around. In simpler words, the recommendation system fetches videos and sorts them in order of which ones a specific user is most likely to view. This, in turn, helps in boosting viewer satisfaction.
In fact, since 2015, viewer satisfaction has been perceived by YouTube as the top-ranking factor In 2012, watch time was the top-ranking factor and in 2011, clicks and views took the cake.
Ranking the videos as per their view and/or click count prompted creators to lure viewers in with the help of misleading titles and thumbnails. As for watch time, it was bumped down the list of the top-ranking factors after it was found out that this approach wasn’t good enough to tell if the viewers actually enjoyed the content they watched. Therefore, the decision to go with viewer satisfaction was taken.
What YouTube does is that it uses surveys and engagement signals to learn a user’s preferences. Engagement signals include metrics such as the likes, dislikes, clicks on the “not interested button,” average view duration, and average percentage viewed.
When it comes to the home page ranking algorithm, YouTube ensures that the most relevant suggestions are presented to each user. This is ensured via:
- Performance: YouTube takes the click-through rate along with average view duration into account to assess how well of a job a video did to retain the viewers’ attention when they spotted it on their home page.
- Personalization: YouTube relies on a user’s watch history, as well as, how frequently they check out a particular topic or channel. This enables it to personalize the said user’s recommendations.
Another tip to retain viewers’ interest is by following a regular uploading schedule.
Now, let’s shed some light on the “suggested” section i.e. the list of videos on the screen’s right-hand portion while another one is playing. This section gives the user many options to choose their next clip to view.
Not only does the said section display videos that are similar to the one playing on the screen but also the ones relevant to the content the user was watching prior to that.
The suggested videos algorithm ranks content by taking the following signals into account:
- Previously-watched videos
- Videos frequently watched together
- Videos falling under the same topic
Moreover, urging viewers to check out another video can prove to be helpful too. Last but not least, using playlists and end screens may increase the chances of your videos making it to the suggested section.
But here comes yet another twist i.e. external factors. One thing marketers should take note of is that external factors can affect their content’s rankings on both the home page as well as in the suggested section. And there’s barely anything that a creator can do about this hurdle.
Among the notable and uncontrollable external factors, the top ones are:
- Interest in a Topic: The interest level in a topic depends on the audience for it. While certain topics amass widespread interest, others are niche in nature and can only pull in a specific type of audience.
- Competition: If your audience is more busy watching videos from other channels, this might affect the ranking of your content in their recommendation list.
- Varying Interests: YouTube users’ interests change by the season, so it’s highly possible that a particular topic popular in the winter of 2020 would go out of style by the summer of 2021.
So, by accepting the fact that people’s interests change frequently, you may turn your attention to a different topic and see if that gets the numbers up.
Sources: How YouTube Works / Creator Insider - Behind the Algorithms / CI video 1 / CI video 2 / CI video 3.
Read next: Report Reveals Insights to Uplift Your Brand Reach and Audience On Instagram as The Marketing Strategy On the Social Network Changes in 2021