[FAQ]

Hey everyone,

I’m creating content for TikTok and I’m really trying to understand how videos get discovered and pushed out to a wide audience. It sometimes feels like magic, but I know there’s an algorithm behind it.

What are the main factors that the “For You Page” (FYP) algorithm looks at in 2025 to decide which videos to show to which users? I’m curious about the importance of things like hashtags, trending sounds, how long people watch a video, and other user interactions.

Any insights into how this system works would be amazing, thanks!

[Jeff Bullas]

The “For You” page algorithm is the core of the TikTok experience, designed to serve each user a highly personalised and continuous stream of content to keep them engaged. While its exact formula is a closely guarded secret, TikTok has shared insight into the key signals it prioritises.

First, user interactions are an extremely strong set of signals. The algorithm learns from almost everything you do on the platform. This includes the videos you like, the videos you share, the accounts you choose to follow, the comments you post, and even the creators or types of content you indicate you are not interested in by hiding them.

Second, video information provides essential context for the algorithm to categorise your content. This includes the details in your video’s caption, the sounds or music you choose, and the hashtags you use. Utilising a trending sound or relevant, specific hashtags can help the algorithm understand your video’s topic and show it to people who have shown interest in similar content.

Third, device and account settings also play a part, although they are generally considered to be weighted less heavily than direct interaction and video information signals. This can include your language preference, country setting, and device type. These signals are mainly used to optimise the performance of the system and ensure content is relevant to your region, but they are not considered core content preference indicators.

Within user interactions, certain signals are understood to be more powerful than others. A very strong indicator of interest is the video completion rate – whether a user watches your video all the way to the end. An even stronger signal is a rewatch. Shares are also a powerful signal, as they indicate a user found the content valuable enough to pass on to others.

When you first publish a video, the algorithm typically shows it to a small initial batch of users. It then measures how this test group interacts with it. If the video performs well with this group, demonstrating strong engagement signals like a high completion rate, the algorithm will then push it out to a larger and larger audience.

It is also worth noting that TikTok has stated that metrics like the number of followers an account has or the performance of previous videos are not direct factors in the recommendation system for any single video. This means that any video from any account has the potential to be discovered and go viral if it resonates strongly with viewers.

Cheers,

Jeff

[Author]

Posts