The TikTok algorithm is designed around two things: getting users to stay and keep them coming back. That’s what a report from The New York Times says, which reviewed a leaked copy of an internal TikTok document that outlines how the system works. The report offers a rare look at one of the most discussed algorithms in technology right now. Next, we will know more details about it. 

The TikTok secret

According to The New York Times, to keep users watching and coming back, TikTok has four main goals in mind. First, value for the user. Second, long-term user value. Third, value to the creator. Lastly, value to the platform. In fact, one of the ways this is done is that the TikTok algorithm prioritizes content diversity. Instead of overwhelming users with a single topic that they might like.

The reason for this is that if the application only offers one type of content, people quickly get bored. So to avoid this, the social network shows a “forced recommendation” to present something new. In that sense, the leaked document presents a simplified version of how the TikTok algorithm works. About what people like and what should be reproduced.

How TikTok Recommends Content

Broadly speaking, it boils down to a combination of likes, comments, video viewing time and whether it has been played, according to the report. Also, it focuses on the creators when it comes to judging the value of their feed. The reason is that it analyzes the “quality of creation” by the rate of publication, retention and monetization that influencers have. However, a spokesperson noted that influencers’ monetary earnings are not important to the TikTok algorithm.

In reality, the transnational company has always shown itself as a company open to information. In its blog posts, the company has detailed the basics of how its algorithm works. Likewise, in their “Center for Transparency and Responsibility”, they address concerns about maintaining security on the platform.


TikTok claims that it takes a number of engagement signals into account when determining what to show people. He has also emphasized that he always invests to customize content preferences. And automatically skip videos that are not relevant or age appropriate for users. This always in relation to the guidelines of the community. 

Mainly, the leak offers a fascinating insight into how the TikTok algorithm works. The details presented here, such as the recommendation formula, are “greatly simplified” according to what is indicated in the document. However, it indicates that social platforms should base their algorithm on clear objectives. Above all, to benefit digital marketing professionals and community managers.