New research highlights how TikTok’s short video format allows the platform to collect user data up to 600 times faster than YouTube, turbocharging its famous recommendation algorithm, Reuters reports.

Published last month by researchers from the U.S. and Germany, the study examines data from 347 TikTok user accounts and five automated bots. It finds that the platform’s algorithm “exploits the user interests in 30% to 50% of the recommendation videos” as it rapidly gathers insights.

“Because it’s in bite-size format, it is a short video, you’re able to collect data about a user’s preference a lot faster than YouTube, where maybe the average video is just less than 10 minutes long,” Jason Fung, former Head of TikTok’s Gaming Unit, told Reuters. “Imagine you’re collecting data about a user on average every 10 minutes versus every couple seconds.”

The researchers write that this approach of frequently recommending videos outside a user’s known interests aims to “either infer better the user interests or maximize user retention.”

According to Reuters, TikTok’s advantage stems from its mobile-first design and early entry into short video, which provides years of data and development experience compared to rivals like Instagram Reels and YouTube Shorts. Its interest-based algorithms differ from the social graphs used by Meta.

“Their recommender system is very common. But what really distinguishes TikTok as an app is the design and the content,” Utrecht University associate professor Catalina Goanta stated, Reuters reports.

The rapid data collection is part of why TikTok’s Chinese parent ByteDance deems the algorithms core to operations. Sources tell Reuters ByteDance would rather shut down the app than sell it amid U.S. security concerns.

Ari Lightman, a Carnegie Mellon professor, said another TikTok tactic, encouraging public user groups via hashtags, allows for “effective learning about its users’ behavior, interest, alignment and ideology.” While U.S. giants could replicate TikTok’s technology, Lightman said the bigger challenge may be replicating “the user culture enabled by TikTok.”

One former ByteDance director, Yikai Li, credits cheap labor in China for its ability to manually tag huge volumes of user data to train its AI, which gives it an edge over U.S. rivals.

“It’s a lot of work sorting out these tags. It’s very laborious. So, Chinese companies have an advantage here. You can afford a lot more people. The cost is cheaper than it is for North American companies,” he said, Reuters reports.