The YouTube Algorithm in 2022 and its Technical Aspects

Information in this post was curated from Hootsuite, Reddit, and my previous blog post on the topic.

Image courtesy of Unsplash

Hold on a Minute and Remember This.

Based on the information provided in my previous blog post that provides general info on YouTube's algorithm-based recommendation systems and their corresponding factors. I am excited to dive deep into the technical aspects of these algorithms and how their factors affect content on the platform. However, it is essential to recognize that technical data on YouTube's algorithm is mainly speculative by nature. Since YouTube does not publicly share detailed information on how its algorithms work, content creators have emphasized anecdotal evidence to gather information for this blog post.

Before we begin, let's recap the goals of YouTube's algorithms, upon which YouTube promotes videos according to Hootsuite:

1. Finding the right video for each viewer.

2. Enticing them to keep watching.

These goals are achieved by YouTube using the following algorithms:

1. The homepage recommendation system which also extends to the videos recommended after watching one video.

2. The systems ranking search results for YouTube videos

Image courtesy of Unsplash

The Homepage Algorithm

The previous goals mentioned are achieved with consideration to:

1. Personalization: the viewer's history and preferences

2. Performance: the video's success in terms of engagement, appeal, and satisfaction.

3. External factors: the video's overall audience and market in terms of seasonality, topic interest, and competition.

Measuring performance metrics is one vector of how YouTube determines which videos appear on its homepage. These metrics include:

1. Click-through rates: the rate at which viewers click on a video.

2. Average view duration: the duration of time which viewers watched a video for.

3. Likes: the metric collected from viewers rating their satisfaction with a video

4. dislikes: the metric collected from viewers rating their dissatisfaction with a video.

5. Viewer surveys: recently introduced, these surveys appear on videos that are gaining steam, often determining whether or not YouTube should continue to promote a video or channel.

As mentioned previously, YouTube also considers personalization factors that recommend videos based on a viewer's relevant interests based on their past behaviors on the platform. Say I have been mostly watching YouTube content on the popular game, Elden Ring. Based on this fact, YouTube will continue to recommend me Elden Ring videos on its homepage that positively correlate with the performance metrics mentioned above.

How YouTube Determines Which Videos We Should Watch Next!

After YouTube builds an interest profile for a viewer using performance and personalization, the algorithm is most likely to recommend:

1. Videos that are often watched subsequent to each other by other viewers of the same interest profile.

2. Videos related to the one just viewed by topic.

3. The videos a viewer has watched in the past.

Image courtesy of Unsplash

How YouTube determines its search algorithm!

Switching gears from the previous algorithmic system aimed at suggesting the videos that support its two goals, YouTube's search algorithm operates differently.

The most crucial factor YouTube uses in its search algorithm is SEO, according to YouTube content creators on Reddit. Refer to the excellent blog posts on the SMMU that explain how you, as a creator, can utilize SEO to have your content reach your target audiences more effectively.

SEO Factors affecting YouTube's search results algorithm:

1. Keywords: YouTube relies on the keywords content creators use in a video's metadata to decide where your video belongs. Meaning that it is important to include the exact keywords that categorize your videos best.

2. Performance: this is a vital step in which YouTube conducts a hypothesis test by showing videos to people on the search results page. This test uses the same performance metrics mentioned above, such as click-through rates, watch times, and viewer satisfaction.

Now that I have explained the technical aspects of YouTube and its algorithms, you can see how interconnected they are, which can be advantageous to your content's creation if you aim to succeed on the platform. Stay tuned to the SMMU for my next blog post explaining how you can use the information in my previous blog on the topic and the data curated in this blog to ultimately build a comprehensive strategy in how you approach making YouTube content.

Resources

Aljebrin, Faisal. “Exploring How YouTube's Algorithm Works in 2022.” The SMMU, 3 Apr. 2022, https://www.thesmmu.com/post/exploring-how-youtubes-algorithm-works.

Cooper, Paige. “How Does the YouTube Algorithm Work in 2021? The Complete Guide.” Social Media Marketing & Management Dashboard, Hootsuite, 21 June 2021, https://blog.hootsuite.com/how-the-youtube-algorithm-works/.

Jain, Rishab. “R/Newtubers - the STB Technique: Cracking the YouTube Algorithm.” Reddit, Reddit, 1 Sept. 2021, https://www.reddit.com/r/NewTubers/comments/pfodr9/the_stb_technique_cracking_the_youtube_algorithm/.