The Secret to Success on YouTube? Kids

Kids love YouTube. They love the pinging of the xylophones in the upbeat “Thank you song” on CoCoMelon, a channel with more than 53 million subscribers that plays animated nursery rhymes. They love watching other kids open and test toys, as they do on Ryan ToysReview (subscriber count: 20,749,585). And they love the Baby Shark song. Possibly because of the fun dance moves and possibly because they want to drive adults crazy.

These trends are nothing new, but now we have more than vast subscriber counts or astounding click numbers to illustrate just how central videos featuring kids are to the platform. In a report Thursday, the Pew Research Center said that in the vast ecosystem of YouTube’s English-language videos, children’s content and content featuring kids under 13 are some of the most popular videos on the site.

For the study, researchers analyzed the videos posted by 43,000 YouTube channels, each with more than 250,000 subscribers, during the first week of 2019. There was a lot to work with. In those seven days, these channels posted almost a quarter-million videos totalling more than 48,000 hours. For the record, the authors note, “a single person watching videos for eight hours a day (with no breaks or days off) would need more than 16 years to watch all the content.”

Those videos covered everything from politics to video games. Most were not intended for kids. But the most popular featured kids. Researchers found that just 2 percent of the videos they analyzed featured a child or children that appeared to be younger than 13. “However, this small subset of videos averaged three times as many views as did other types of videos,” says the report.

There have been studies of niche communities within YouTube, but “We hadn’t seen something like this done before,” says Aaron Smith, director of the data lab team at Pew. Although YouTube children’s content wasn’t the impetus for the study, Smith says the results weren’t surprising: “We had a sense that this kind of content would be fairly popular. We know that lots of parents let their kids watch videos on YouTube.”

Videos with cheery, if nonsensical, titles like “Funny Uncle John Pretend Play w/ Pizza Food Kitchen Restaurant Cooking Kids Toys,” and “No No, Baby Rides the Scooter!” racked up over 6 million views each. “SUPERHERO BABIES MAKE A GINGERBREAD HOUSE SUPERHERO BABIES PLAY DOH CARTOONS FOR KIDS,” attracted almost 14 million views.

Not all the videos that featured young kids were nursery rhymes or traditional kids content like Mister Rogers’ Neighborhood. Pew’s analysis found that only 21 percent of videos featuring children appeared to have been aimed at kids. But videos that were both aimed at kids and featured kids were the most popular videos in Pew’s analysis, averaging four times as many views as other “general audience” videos.

As for the other 79 percent of videos that had kids but weren’t directly aimed at children? They did better too, getting “substantially more” attention than other videos aimed at teens and adults. The five most popular videos from the week Pew studied included a baby name reveal and family vlogs with titles like “WELCOMING A NEW MEMBER OF FAMILY!!” One was a sliming video. None are immediately alarming, though Smith couldn’t comment on why that kind of content was so attractive to so many viewers. “Why that type of material pops is unclear to me,” he says. “Someone clearly is enjoying it but it’s not clear who those folks are or what their motivations are for doing that.”

YouTube insists that the platform is not intended for children. “We can’t speak to Pew’s methodology or results. But generally on YouTube, the most popular video categories tend to be areas like comedy, music, sports and ‘how to’. And we have always been clear YouTube has never been for people under 13,” Ivy Choi, a YouTube spokesperson, says via email. In 2015 YouTube created YouTube Kids, a platform aimed at children that is supposed to be safer and easier for parents to control and monitor. Pew’s report only examined YouTube’s main platform and YouTube’s API doesn’t provide information about whether views come from the YouTube site or YouTube Kids. However, as YouTube explains, all YouTube Kids content is also available on the main YouTube site.

While YouTube’s main platform may not be intended for children, Pew’s report shows that children are extremely popular on the site. With that popularity come concerns. Pedophiles have used the platform to find videos of partially clothed children and tell others where to look. A report from the New York Times found that YouTube’s recommendation algorithm suggested videos of children to pedophiles. Last week, the Federal Trade Commission reportedly agreed on a multi-billion dollar settlement with YouTube for failing to comply with the Children’s Online Privacy Protection Act, which prevents companies from tracking information and data about children younger than 13. There are also questions about who is making money on the content that is aimed at young audiences. The creators behind the ever-popular CoCoMelon are unknown, as are most of the other top creators of children’s content. Neither Wired nor The Wall Street Journal were able to identify the owners of nine of YouTube Kids’ top ten channels.

Pew’s study doesn’t include any information about who is watching the videos, or why one type of content is more popular than another. The report found that cross-posting on other social media sites is linked to an increase in views. But there are other factors that could drive popularity. A video may catch on because it is highly ranked by YouTube’s recommendation engine, or because it was featured on a popular channel. Or it may just go viral. But YouTube’s API doesn’t give researchers that kind of insight. Understanding the causes is “opaque and almost impossible to uncover without being behind the scenes,” says Smith.

Smith says the Pew report is far from exhaustive, but is a “useful first step.” The study also illustrates just how complicated studying YouTube can be. On the one hand, lots of information about who is watching videos, how they become popular, and which channels are monetized, isn’t provided by the API. On the other hand, YouTube is so large that it was challenging to evaluate as a whole. Smith’s team didn’t attempt to look at non-English content, which makes up the vast majority of videos on the site. They also only studied content from a limited period of time. But even in just one week, they found themselves dealing with hundreds of thousands of videos. “You can extrapolate out from that to estimate just how big YouTube is,” says Smith. “It’s almost unfathomable.”