Overview
Interest insights help reveal the topics, themes, and content categories that resonate most strongly with users in each audience cluster.
These insights are based on audience behavior and profile analysis rather than self-reported preferences, helping provide a more objective view of audience affinities and engagement patterns. Interest analysis is one of the core components of TikTok Insights reports and plays a central role in audience segmentation and audience understanding.
How Interests Are Determined
Interests are inferred by analyzing TikTok post content and profile biography information This allows Audiense to identify the topics and categories most strongly associated with audience behavior and self-expression. Because the analysis is based on actual audience activity and profile-level signals, interest insights provide a strong indicator of genuine audience affinity.
Understanding the Interest Taxonomy
TikTok Insights organizes interests using a two-level taxonomy structure.
The taxonomy includes:
- 25 parent categories
- 498 specific interests
For example, a broader category such as Sports may contain more specific interests such as Soccer, while Music may contain interests such as K-Pop. This structure allows organizations to analyze interests at both broad and highly granular levels. Users may belong to multiple interests and categories simultaneously.
Understanding TML (Times More Likely)
TML stands for Times More Likely. TML measures how much more or less likely a cluster is to engage with a topic compared to a comparison baseline audience.
For example:
- 1.0x indicates alignment with the baseline audience
- 2.0x indicates the cluster is twice as likely to engage with the topic
- 0.5x indicates below-average affinity
By default, the comparison baseline is set to Global - General, although other comparison groups may also be supported. TML helps organizations identify which interests are most distinctive for a cluster rather than simply the most common.
Interest Visualization Options
TikTok Insights currently supports two primary interest visualization modes. These visualizations help users explore audience affinities from both high-level and detailed perspectives.
Treemap View:
The treemap view displays interest categories as a grid of colored rectangles. The size of each rectangle reflects the penetration score for that category within the cluster. Larger rectangles represent categories associated with a larger percentage of the audience.
At the category level, colors act as fixed category identifiers. When drilling into a category, color intensity reflects TML affinity relative to the comparison baseline. Treemap visualizations are useful for quickly identifying dominant audience interests and major affinity areas across a cluster.
Bar Chart View:
The bar chart view displays interests as ranked rows alongside additional metrics.
This view includes:
- Penetration percentage
- TML affinity score
- Number of subcategories
The bar chart view also supports grouping interests by category or displaying all interests in a flat ranked list. This view is useful for more detailed interest comparisons and affinity analysis workflows.
Filtering and Sorting Interests
TikTok Insights supports filtering and sorting within interest visualizations. Users can filter interests using keyword searches and sort results using dimensions such as:
- Percentage
- TML
- Subcategory count
Sorting by penetration helps identify the most common audience interests, while sorting by TML helps surface the most distinctive affinities for the clusters.
Topics
The Topics section highlights the subjects your audience posts about most frequently on TikTok and indicates how strongly each topic is associated with the selected audience or cluster. Unlike Interests, which group users into predefined categories based on inferred affinities, Topics surface the actual themes, keywords, and subjects appearing in audience conversations. This provides additional context about the content your audience is actively creating and engaging with. Each topic includes several metrics to help evaluate its significance within the audience.
Topic Metrics
The Topics table includes the following information:
-
Topic: The keyword, phrase, or subject identified within audience posts.
-
Percentage (%): The percentage of audience members associated with the topic. Higher percentages indicate topics discussed by a larger portion of the selected audience or cluster.
-
TML (Times More Likely): Measures how much more likely the selected audience is to discuss the topic compared to the selected comparison audience. Higher TML values indicate stronger affinity for that topic.
-
Relevance: A qualitative indicator that summarizes how distinctive the topic is for the selected audience relative to the comparison baseline. Topics with high TML values are typically identified as Distinctive, helping surface the subjects that make the audience unique.
Comparison Audience:
The Topics analysis compares the selected audience against a baseline audience, which is displayed at the bottom of the report. By default, the comparison audience is Global, General, allowing you to identify the topics that are most distinctive when compared to the broader TikTok population. Other comparison audiences may be available depending on your report configuration.
Using Topics for Audience Analysis:
The Topics section helps you understand not only what your audience is interested in, but also what they are actively talking about. Reviewing the most distinctive topics can help identify emerging trends, campaign opportunities, niche communities, and the language your audience naturally uses when discussing products, brands, or interests.
