An analysis of 1,100 TikTok users' watch histories across six months in 2024 shows just how effective TikTok is at getting even its heaviest users to watch more
More than 800 U.S. TikTok users shared their data with The Washington Post. We used it to find out why some people become power users, spending hours per day scrolling.
Context & Ripple Effects
This watch-history analysis adds behavioral evidence to a long-running account of TikTok’s engagement advantage. Earlier reporting described an engineering approach centered on retention and time spent, while a separate comparison found TikTok viewers spending longer in the app than Facebook and Instagram combined as viewing time shifted toward TikTok.
It also distinguishes consumption from creation: prior survey data found public posting highly concentrated among a small share of U.S. users even as most users rarely posted. The new analysis focuses on how the recommendation-led viewing experience deepens use among those already spending the most time scrolling.
First-order effects
- TikTok gains evidence that its viewing experience can increase watch time even for heavy users, making its most engaged audience more valuable to the platform.
- Heavy users face a stronger pull toward longer viewing sessions, according to the six-month watch-history analysis.
Second-order effects
- More time spent gives TikTok a stronger engagement benchmark against other consumer video platforms, increasing pressure on rivals to improve recommendation quality and session retention.
- A larger concentration of attention among power users can make audience access more dependent on TikTok’s feed, rather than on users actively seeking out individual creators or posts.
Third-order effects
- If repeated across broader samples, this pattern would reinforce a platform model in which recommendation systems—not posting rates—are the primary mechanism for capturing user attention and gatekeeper leverage.
- Evidence that engagement rises even among already-heavy users could sharpen scrutiny of design choices optimized for time spent, though this analysis alone does not establish causation or harm.
The trend: Consumer video platforms are competing less on social posting and more on recommendation systems that continuously expand time spent by their most engaged viewers.