Over 1 billion hours of YouTube videos are viewed every day by people around the globe. This includes everything from videos for cats to cat videos. The YouTube algorithm is the recommendation system that decides which videos YouTube suggests to those 2 billion-plus human users (and untold numbers of feline users).
For marketers, influencers, creators and others alike, this raises a fundamental question: How do you get YouTube to recommend your videos, helping you earn more likes?
This blog post will explain the algorithm and the latest changes to 2021. We’ll also discuss the most recent changes and show how the pros use YouTube’s search system and discovery system to get videos in front people’s eyes.
Bonus: Get the 30-day free plan to quickly grow your YouTube audienceA daily workbook with challenges to help you get your YouTube channel growing and measure your success. Real results in one month
An overview of YouTube’s algorithm
Let’s first look at how “algorithms” became so ubiquitous in our daily lives.
2005-2011: Optimizing clicks & views
Jawed Karim, a.k.a. founder of YouTube, said that he was inspired by Jawed Jackson and Justin Timberlake’s Superbowl performances. Jawed Karim (a.k.a. Me at the Zoo) is the founder of YouTube. He created it in 2005 to crowdsource video of Janet Jackson’s Superbowl performance and Justin Timberlake. It is not surprising that YouTube recommended videos with the highest views and clicks for years.
This led to the proliferation of misleading thumbnails and titles, or in other words,clickbait. As videos made people feel unsatisfied or tricked, the user experience plummeted. Getfans.io can get subscriber boost for your channel.
2012: Optimizing to Watch Time
In 2012, YouTube adjusted its recommendation system to support time spent watching each video, as well as time spent on the platform overall. People watch videos longer if they find them valuable or interesting, according to theory.
Some creators tried to make videos shorter to increase their chances of viewers watching them to the end, while others made longer videos to increase their overall watch time. YouTube did not endorse these strategies and maintained its party line: Make videos that your audience enjoys watching and the algorithm will reward them.
However, quality time is more important than time spent on the internet. YouTube has changed their approach.
2015-2016: Optimizing to Satisfy
YouTube started measuring viewer satisfaction with surveys in 2015. It also prioritized direct response metrics such as Shares, Likes, and Dislikes.
In 2016, YouTube released a whitepaper describing some of the inner workings of its AI: Deep Neural Networks for YouTube Recommendations.
The algorithm was now much more personal. The goal was to find and capture the perfect video.Each viewer has a preference about what they want to seeIt doesn’t just include the video you may have seen in the past.
In 2018, YouTube’s Chief Product Officer mentioned that he was the reason.70% of YouTube users spend more time watching videos than they do on YouTube.Watch videos that the algorithm recommends.
2016-present: Brand safety, demonetization and dangerous content
YouTube’s popularity and size over the years has led to an increase in content moderation issues. What the algorithm recommends has become a major topic for not only creators but also government officials and news media.
YouTube stated that it takes responsibility for supporting diverse opinions and reducing misinformation. Algorithm changes enacted in early 2019, for example, have reduced consumption of borderline content by 70%. (YouTube defines borderline content as content that doesn’t quite violate community guidelines but is harmful or misleading. Violative content, on the other hand, is immediately removed.)
Creators who are afraid of accidentally breaking ever-changing community guidelines could be punished with strikes, demonetization or worse. (And in fact, according to CEO Susan Wojcicki, one of YouTube’s priorities for 2021 is increasing transparency for community guidelines for creators). This also impacts advertisers and brands, who don’t want their logo and name to be associated with white supremacists.
American politicians are becoming more concerned about the social role of social media algorithms such as YouTube’s. YouTube (and other platforms) have been summoned to account for their algorithms at Senate hearings, and in early 2021 Democrats introduced a “Protecting Americans from Dangerous Algorithms Act.”
Let’s now discuss what we know so far about the workings of this deadly beast.
YouTube’s algorithm selects videos with two goals in view:Finding the perfect video for every viewerAndYou can encourage them to continue watching.
The algorithm refers to three different, but related, selection or discovery systems.
One that selects videos to be used in theYouTube homepage;
One that ranks the results for any givenSearch; and
One that choosesVideo suggestionsFor viewers to view next.
YouTube says that in 2021, homepage and suggested videos are usually the top sources of traffic for most channels. Search is the best source of traffic for instructional or explainer videos (e.g. “How to tune up your bicycle”) which are often seen less than those that have more content.
YouTube decides which algorithm to use
YouTube uses which ranking signals to decide which videos it will show people.
Every traffic source is different. However, the mix of these factors can affect your video’s views count.
- personalization(the viewer’s preferences and history)
- Performance(The video’s success).
- External factors(The overall market or audience)
YouTube’s homepage algorithm
YouTube offers a wide variety of videos to choose from when a user opens YouTube or types in youtube.com.
The algorithm is still trying to figure out what the viewer desires: acoustic pop song covers. Inspirational speeches against procrastination? To catch up on their favorite possum blogger?
Two types of ranking signals determine which videos are selected for the homepage:
YouTube uses metrics to measure its performance, such asClick-through rate, average view length, average percentage viewed and likes/dislikesAnd viewer surveys. After you upload a video, the algorithm shows it to a few users. If it is appealing to, engages and satisfies them (i.e. they click on it and watch it through all the way, like it and share it, etc.). The video is then offered to more viewers via their homepages.
YouTube isn’t a popular tab. YouTube uses personalization to offer videos to users it believes are interested.Based on their past behavior, relevant to their interestsa.k.a. Watch history. You will see more of what you like if you watch a lot of the same channel or a certain topic. This is because of the fact that it can detect changes in behaviour as interests and affinities change over time.
YouTube’s suggested video algorithm
YouTube uses slightly different criteria when suggesting videos for people next. The algorithm learns more about what people are interested in by watching a few videos. It then suggests some options to the right.
This algorithm, in addition to personalization and performance, is more likely to recommend:
- Video clips that are often viewed together
- Videos that are topically related
- Videos that the user has seen in the past
Pro Tip:YouTube Analytics can be used by creators to see what your audience has seen. This will help you narrow in on the broader topics or interests that your audience is interested in.
Pro Tip #2:A successful technique is to make a sequel. Ryan HigaA video on singing technique went viralHe didn’t release the sequel until three more years later. But timing is your choice.
YouTube’s search algorithm
YouTube is both a search engine and a video platform. This means that SEO knowledge is essential.
Sure, sometimes people go to YouTube to hunt down a specific video to watch (hello again, peanut butter baby). Even then, the algorithm determines how to rank search results when you type “peanut butter babies”
How can you rank your video near the top of searches?
The keywords in your video’s metadata determine what Youtube’s search algorithm will look for. These two words will help your video show up in searches for videos related to laparoscopic surgery. We have more keyword advice, so continue reading.
PerformanceOnce the algorithm has determined what your video is it will show it to search engines. This is where performance (click-through rates, watch time and survey feedback, etc.) becomes important. It is crucial to pay attention to performance. Your video should appeal to people searching for your keywords. It will be seen by more people and rise up the SERPs.