Here is a question: how does a video site know what to show you next? The answer is a recommendation system. A recommendation system picks videos for you from a huge number of videos, and it suggests videos similar to ones a person watched. Watch history lets a video site give relevant picks. Recommendation systems learn from what people do, such as what they watch and click on. As of August 2026, Common Sense Media says YouTube's algorithm plays a big role in what kids and teens see. Recommended videos are based on what a child watches and searches for. Let's find out how it works, one piece at a time. Here we go, step by step.

Next, look at how the choosing is split into steps, one after another. Choosing from a huge pile takes more than one step. In one common recommendation design, the system starts from a potentially huge collection and generates a much smaller subset of candidates. Candidate generation is the first stage of recommendation: given a query, the system generates a set of relevant candidates. A given recommendation model may provide multiple candidate generators, each nominating a different subset of candidates. The model that generates candidates needs to evaluate queries quickly given the enormous size of the corpus. Then the engagement prediction and ranking algorithm is applied only to a smaller set of candidate posts. Items that appear very rarely may not be updated frequently during training.

Two more steps follow, and then a way to compare. After candidates come scoring and re-ranking. In a scoring stage, the system can use a more precise model relying on additional queries. In a re-ranking stage, the system can remove items the user explicitly disliked or boost the score of fresher content. If popularity is captured carelessly, popular items may end up dominating the recommendations. Systems also differ in what they compare. Content-based recommending uses similarity between items to recommend items similar to what the user likes; Google's course notes give an example: if user A watches two cute cat videos, the system can recommend cute animal videos to that user. Re-ranking can also help ensure diversity, freshness and fairness.

Another method looks at people instead of items. In collaborative filtering, if user A is similar to user B and user B likes video 1, the system can recommend video 1 to user A. Underneath every method sits behavior. In a Columbia Knight Institute explainer, engagement means any score defined only in terms of the moment-to-moment actions of the user, and the explainer says the behavioral record is the fuel of the recommendation engine. To a first approximation, generating a user's feed is a matter of ranking all the posts that can be shown. For every post in a user's feed, the algorithm receives feedback about whether and how the user engaged with it. Recommendation algorithms are largely limited to looking for patterns in behavioral data.

Now look at what YouTube said, and when it said it. Next, here is what YouTube said about itself, with the dates it gave. In a September 2021 post, YouTube said its system sorts through billions of videos to recommend content tailored to a viewer's specific interests. In 2021 YouTube said its system learns every day from over 80 billion pieces of information it calls signals. The importance of each signal depends on the viewer, YouTube said, and its system does not follow a set formula but develops dynamically as a viewer's habits change. YouTube said clicking on a video is a strong sign that a viewer will also find it satisfying. YouTube also said it learned in 2011 that clicking on a video does not mean the viewer actually watched it.

Next, look at what the site tries to predict, with the dates and the numbers that go with it, one by one. Before 2012, YouTube optimized for click-through rate instead of watch time, which the Knight Institute says led to clickbait thumbnails. YouTube added watch time to its recommendations in 2012, and YouTube said that when it first added watch time to recommendations it saw an immediate 20% drop in views. Next, as of November 2026, the Knight Institute explainer says YouTube optimizes for expected watch time: how long the algorithm predicts the video will be watched. As of November 2026, it also says that if a user sees a video in their recommendations and does not click on it, the watch time counts as zero.

As of November 2026, the Knight Institute explainer says Netflix originally optimized for suggesting movies that the user is likely to rate highly on a scale of one to five. As of November 2026, Common Sense Media says YouTube's business model rewards watch time: the longer users watch, the more YouTube makes in ad sales. Ratings and reactions are signals too. YouTube said in September 2021 that it asks why a viewer gave a video a one- or two-star rating. It said only videos rated four or five stars are counted as valued watch time. YouTube said people are, on average, more likely to be satisfied by videos that they share or like, and that a dislike is a signal that the viewer probably did not enjoy watching the video.

Next, look at what the site does with your time. Here is the next question: what are these systems built to do, and who are they built for? As of August 2026, Common Sense Media says the YouTube algorithm isn't designed with kids in mind and is designed to keep users engaged; it also says YouTube's Up Next feature and autoplay encourage hours of uninterrupted viewing, often leading kids to lose track of time. Features such as autoplay are designed to keep kids watching, and autoplay and endless scrolling can make it hard to stop watching. As of November 2026, KidsHealth says most apps, games and streaming platforms are designed to keep users involved as long as possible.

Age rules come next, with the facts dated November 2026 unless noted. YouTube's terms of service say you must be at least 13 years old to use YouTube, though children of all ages may use YouTube and YouTube Kids if a parent or guardian enables it. Common Sense Media says YouTube requires users to be at least 13 years old to create an account; in most countries, 13 is the minimum age to manage your own Google Account, and for children under the minimum age, parents can help create and manage a Google Account with Family Link. In a 2019 news report, a YouTube spokesperson said YouTube is not a site for people under 13. Congress enacted the Children's Online Privacy Protection Act (COPPA) in 1998. Websites covered by COPPA must notify parents and get their approval before collecting a child's personal information.

YouTube Kids and viewer controls come last, as of November 2026. In YouTube Kids, recommended videos are also based on the content setting a parent selected. Controls made for kids are not perfect, and a child may still see something a parent did not want. On certain pages, such as Home and Watch Next, a viewer can mark a video as Not interested. YouTube says that if a viewer deletes watch history, it won't base future video recommendations on that content. If Autoplay is on and a viewer wants to stop the next video, they can tap Cancel at the end of a video. On the Time management tab, a YouTube viewer can change the Take a break reminder and Auto-play settings. KidsHealth says to turn off autoplay on streaming platforms.

Words to know

• algorithm: step-by-step rule
• candidates: possible picks
• autoplay: automatic next video
• signals: pieces of information
• engagement: user actions