TikTok’s valued around $20-$30 billion dollars with more than 800 million active users (100 million monthly active users in US). What makes the product worth $30 billion dollars? Besides the user base, I believe it’s the algorithm. We know now more about the algorithm in Tik Tok’s recent suit filing against the US government.
How does Tik Tok’s Algorithm work?
Let’s dive into how the algorithm works.
Overview of the Algorithm
TikTok’s algorithm uses machine learning to identify what content a user is most likely to engage with. After each engagement, it attempts to serve them more similar content. So, more of videos that are similar or are liked by people with similar user preferences.
User Journey with the Algorithm
Let’s understand how the algorithm works at different stages of the user journey — from initial onboarding to becoming an engaged user;
- Initial User Onboarding – when a user opens TikTok for the first time, they are shown 8 popular videos featuring different trends, music and topics. After that, the algorithm will continue to serve the user new iterations of 8 videos based on which videos the user engages with.
- Ongoing Content Recommendations – beyond initial onboarding, the algorithm identifies videos similar to those that have engaged a user based on video information — location, comments, captions, hashtags or sounds. Furthermore, Tik Tok also considers user device and account settings — language preference, country setting, and device type.
- Generating “Clusters” of videos and Users – once TikTok collects enough data about the user, the app is able to map a user’s preferences in relation to similar users and group them into “clusters.” Simultaneously, it also groups videos into “clusters” based on similar themes, like “basketball” or “bunnies.”
- Engagement within Clusters – with machine learning, the algorithm serves videos to users based on their proximity to other clusters of users and content that they like.
- Avoiding Redundancies – Tik Tok’s algorithm avoids redundancies that could bore the user. For example, seeing multiple videos with the same music or from the same creator.