YouTube Algorithm

YouTube is a video sharing platform where people can watch videos, upload their videos, and comment on uploaded videos. Every kind of video is available on YouTube, such as education, fashion, cooking, travel, etc.

The YouTube algorithm is one of the most sophisticated algorithms. The algorithm track all the actions performed by the user inside the app/ website, such as what kind of videos the user likes to watch. The algorithm works on a real-time feedback loop that will categorize the videos for each user on the basis of their interest and their previous activities.

The algorithm basically depends on the data which it gets from the activities performed by the user. It tracks the time that the user spent on YouTube. It also tracks whether the user likes or dislikes the video and which type of video the user is interested in. By considering all these factors, the YouTube algorithm categorizes the videos for each viewer. Sometimes the videos are also prioritized on the basis of the good comments and the maximum number of likes.

History of YouTube Algorithm

Before 2012: View Count

Before 2012, the YouTube prioritize videos based on the number of views on a video. The more the views on videos, the higher would be its rank. The videos with more number of likes come in the category of popular videos, which will get display on the YouTube homepage of most of the users.

But the user being clever has understood the logic behind the algorithm, and created the problem of click-bait. Users started giving fake likes and comments to make any videos popular that resulted in bad quality content and videos to get prioritized.

2012-2016: view duration and session time

YouTube changed its algorithm in 2012 to overcome the problem of click-bait. The new algorithm prioritizes or ranks the videos on the basis of the view duration (watch time) and session time (session time). This algorithm was also not beneficial as the user also cracked the logic of this algorithm. The logic of the algorithm is that the users have to spend an amount of time on a video so that it creates a view on that video, and the user did the same.

2016: Machine Learning

In 2016, YouTube switched their algorithm to Artificial Intelligence (AI) and machine learning. The technical theorists named this algorithm the misinformation engine and the great radicalize. This algorithm has a tendency to show conspiracy theories, fake news, and disturbing content to users.

2017: Quality

In 2017, YouTube designed its algorithm that focused on video quality. This YouTube algorithm improved the quality of popup news and reduced "inflammatory religious or domineering" content.

2018: Monetization

In 2018, YouTube changed its monetization policy to reduce the number of content creators. Then after few months, CNN reported that it included 300 major brands such as Adidas, Cisco, Hilton, and Amazon, running ads on channels promoting white supremacy, Nazis, pedophilia, and other fringe content. Even though they do not check the "brand-safe" box while purchasing advertising space, the brands were not safe.

2019: Banning borderline content

In 2019, YouTube redesigned its algorithm that banned the borderline content that people used to add irrelevant and misleading content in their videos.

How does the YouTube algorithm work?

The YouTube algorithm works on the past activities of the user and provides more and more videos related to the interest of the user. The YouTube algorithm categorizes the videos based on the YouTube channel the user likes to watch, the types of videos the user watch, likes and dislikes on the videos. There is one factor on which the YouTube algorithm focus most is the engagement time of the user on each video.

The YouTube algorithm works on the past activities of the user and provides more and more videos related to the interest of the user. The YouTube algorithm categorizes the videos based on the YouTube channel the user likes to watch, the types of videos the user watch, likes and dislikes on the videos. There is one factor on which the YouTube algorithm focus most is the engagement time of the user on each video.

• Search
• Home and suggested videos
• Trending
• Subscriptions

Search - There are two major components that affect the ranking of the video that are keywords and relevance. YouTube ranks the video in search on the basis of title, description, the content of the video and how well all these factors match with the user's search. The YouTube algorithm also ranks the search results by considering the number of videos watched by the user on a particular channel.

Home and suggested videos - The YouTube algorithm displays the videos on the YouTube homepage by considering the user interest. The algorithm tracks all the activities performed by the user inside the app. Then categorize the videos on the basis of those previous activities and the engagement time of the user spent on the videos for the YouTube home page. This is the reason why every user has a different YouTube home page and suggested videos.

Trending - The trending page has new and popular videos on it. These videos are classified on the basis of the country of the user. The youtube algorithm ranks their videos on the trending page by considering both the popularity and innovation equally.

The algorithm majorly considers the two factors that are the view count and the rate of view growth for each video while ranking the trending videos.

Subscriptions - The youtube provides a feature of subscription for their users. The users can subscribe to their favorite channel and get all the notifications related to that channel. But the youtube algorithm does not consider the number of subscribers on a channel to give a higher rank.

The youtube algorithm considers the factor called view velocity to rank the video. The view velocity measures the numbers of users who watched the videos regularly when it gets uploaded. The higher the view velocity of the video, the higher the rank of the video. The Youtube algorithm also considers the number of active subscribers while ranking the video.

Factors Affecting the youtube algorithm

There are some factors that affect the youtube algorithm while choosing the video for any user that is given below:

• The type of videos users regularly watches, such as if the user likes to watch fashion videos and spent most of the time watching fashion videos, then the youtube algorithm tries to provide more and more videos related to fashion.
• The amount of time users spent watching a particular type of video.
• How often does any channel upload new videos?
• The session time user spent on youtube
• Engagement of user with the video such as giving like or dislike to any video, giving feedback to the video whether it is positive or negative.

How to rank well with the youtube algorithm?

There are some ways that help to provide good ranking to the youtube videos, which are given below:

• The title of the video should be accurate and appropriate. Also include metadata. Add keywords that are readable and easy to understand.
• The description of the video should be attractive must contain meta-phrases and keywords.
• Try to add accurate subtitles to the video.
• Add the interesting thumbnail which matches the video content. Some people watch videos because of a good thumbnail.
• Include a section in which try to convince a user to subscribe to the channel, because the larger the subscribers, the higher the rank of the video.