You are reading :
|
What is TikTok’s Algorithm?

What is TikTok’s Algorithm?

_122473559_gettyimages-1228117993-594x594

The TikTok algorithm in 2021 determines what shows up in your “For You” feed. The TikTok formula, based on our research, indicates that the videos a user will see in their “For You” feed will change over time based on their viewing preferences and habits.

@jonilarnet Highlight: #Breakfastmeetingwithedem #ghanamusic #Jonilar #Jonilarnet #ghanatiktok #fypシ #GhanaMusic ♬ original sound – Jonilar.net

This is how TikTok defines the “For You” page algorithm:

How Does TikTok’s “For You” Page Algorithm Work?

TikTok has been secretive about its algorithms in the past, which has led to much speculation about the factors that affect how the For You Page (FYP) is populated. That changed in June 2020, with TikTok publishing a blog post detailing how the For You Page recommendations are generated.

According to TikTok, the For You Page’s algorithms are based on the preferences and activity history of individual users. That means no two For You feeds are going to be exactly the same, due to the sheer number of factors and variations involved; as the name implies, these recommendations have been carefully curated for you.

The 3 main factors that influence your TikTok recommendations include:

  • User interactions
  • Video information
  • Device and account settings

 

User Interactions

User activity plays a significant role in determining which videos TikTok shows, with the following heavily influencing your For You feed:

  • Video likes
  • Video shares
  • Accounts followed
  • Comments posted
  • Videos created

 

The algorithm also takes into account how far a user got in the video, and whether or not they watched it in its entirety. Watching a video from beginning to end is a strong sign of interest and carries greater weight in the TikTok algorithm than some of the other contributing factors.

Regularly showing up in users’ feeds goes beyond them simply being a follower. Users will see more content from creators that they engage with than those they simply follow but scroll past.

 

Video Information

This encompasses video attributes and data specific to individual videos, including:

  • Length/runtime
  • Captions
  • Sounds
  • Stickers
  • Hashtags

Hashtags in particular are a major component of which videos are recommended. Any user that watches or creates a video with a specific hashtag will be more likely to see videos that use that hashtag in the future.

Device and Account Settings

TikTok has confirmed that its algorithm takes device settings and account settings into consideration, but clarifies that these factors are not given as much weight as stronger, preference-indicative measures—such as likes and video plays—and that these factors are primarily tracked to optimize app performance.

Examples of device and account settings considered include:

  • Language preference
  • Country setting
  • Device type

What Content Does the TikTok Algorithm NOT Show in FYP?

According to TikTok, to keep your “For You” feed interesting and varied, their recommendation system works to intersperse diverse types of content along with those you already know you love. For example, your For You feed generally won’t show two videos in a row made with the same sound or by the same creator.

Content the TikTok algorithm prevents from appearing in users’ ‘For You’ feeds including:

  • “Not interested” content
  • Duplicate content
  • Seen content
  • Spam
  • Potentially Safety-compromising content

What Doesn’t Influence Your TikTok ‘For You’ Feed?

The number of total followers an account has does not result in their posts surfacing higher, or more often, based on the functionality of the For You algorithm. Naturally, those who have a higher number of followers will have more overall visibility, but their posts are not given preferential treatment based on their popularity. Your track record of high-performing versus low-performing videos also does not affect placement.

This means that a TikTok video from a creator with a high number of followers—and/or with a history of high-performing videos—will not, based on those factors alone, get recommended more often than a video from a newer creator with less followers, and/or lower-performing videos.

Videos are measured and recommended based on their own merits, not the merits of their creators.

The relevant passage in TikTok’s blog post states:

“While a video is likely to receive more views if posted by an account that has more followers, by virtue of that account having built up a larger follower base, neither follower count nor whether the account has had previous high-performing videos are direct factors in the recommendation system.”

How Does TikTok Calculate Effect?

Although TikTok shared some information regarding algorithmic functions, they didn’t share everything, leaving some answers up to anecdotal speculation.

The “Batch Effect” is a popular theory among TikTok creators. In the Batch Effect, TikTok essentially shows your video to different batches of people at different times. The way users react determines whether or not your content gets shared with more people. The more views and interactions your video gets, the bigger—and greater number of— batches see it down the line.

The time lag between different batches can take up to several days. Creators have observed that their video views often “jump” significantly after plateauing. This is speculated to be because the next “batch” has launched and their content has been exposed to more users.

 

Do TikTok Filters and Effects Increase Your Views?

Filters and effects are used to make your videos more compelling and engaging for viewers. TikTok offers creators and brands a wide variety to choose from, and regularly releases fresh, new options. Its vast library of effects and filters is a big part of what sets it apart from competitors like Instagram Reels.

Some analysts have noted that TikTok appears to give greater priority to videos that use effects—specifically, effects that are currently trending.