TL;DR
- TikTok’s newsroom, in a post dated 18 June 2020, says neither follower count nor whether an account has had earlier high-performing videos is a direct factor in its recommendation system.
- The same post says a video from an account with more followers is likely to get more views, because that account has built a larger base.
- OpenSponsorship’s follower-tier analysis of 1,502 paid deliverables shows the second statement holding at the extremes: median views are 1,115 for nano creators and 20,123 for mega. It also shows views per follower falling from 0.199 for nano creators to 0.011 for mega.
- In our pilot campaign 113 of 698 videos passed 10,000 views and brought 83.3% of the views (deck figures, not independently audited).
- A creator’s history shows who their audience is and what they have made. TikTok says past hits are not a direct ranking factor, so history alone cannot tell a brand how far the next video will travel.
TikTok’s newsroom names three groups among the factors behind its recommendations, and follower count is not one of them. The post says outright that neither follower count nor earlier high-performing videos is a direct factor.
TikTok names three groups of factors, and follower count is not among them
In its newsroom post of 18 June 2020, TikTok says recommendations rest on user interactions such as likes, shares, comments and follows, on video information such as captions, sounds and hashtags, and on device and account settings such as language, country and device type. It adds that a strong indicator of interest, such as finishing a longer video, weighs more than a weak one.
The page is more than six years old, so it describes the system as TikTok described it then.
A bigger account still gets more views on average
TikTok does not claim followers make no difference. The same sentence says a video is likely to receive more views if an account has more followers, because that account has built up a larger base. The difference is in the mechanism. Followers are viewers a creator already has. They are not a score the system adds to the ranking.
OpenSponsorship’s follower-tier data, from 1,502 paid deliverables created between January 2025 and June 2026 on its own platform, shows both effects together. Median views per deliverable are 1,115 for nano creators and 20,123 for mega, so the largest accounts get more views on average. Views per follower fall from 0.199 for nano creators to 0.011 for mega, so a small account’s video travels further relative to its audience.
An earlier hit does not set the next video’s reach
TikTok names previous high-performing videos in the same sentence as follower count, and says they are not a direct factor either. That matters to anyone treating a creator’s past results as a forecast for the next post.
In our pilot campaign (deck figures, not independently audited), 113 of 698 videos passed 10,000 views and brought 83.3% of the views, and 301 creators posted the 698. A brand that chose by past performance would have been choosing the audience a creator already has. TikTok’s own account says videos are ranked by each viewer’s likely interest. Researchers who tried to predict how far posts spread on Twitter found that the best models explained less than half of the variation, which is the same difficulty on another platform.
Choose the person who can make the post
Read recent posts for evidence of language, format and audience fit. Check whether the person can film during the launch window. A previous hit can prompt a closer look, but the new post still needs a creative reason to exist. See how a creator shortlist is reviewed.