The goal of the Language field is to identify the language a creator publishes in, not every language they may speak.
A creator may have an English bio and an Italian display name, but if their content is consistently in Spanish, Spanish is the language relevant to a brand.
Text, audio, and metadata signals are combined to detect the primary publishing language.
Coverage
Share of profiles where we resolved a confident primary language.
Coverage by Platform
How Detection Works
The system analyzes multiple content signals from a creator's public profile to determine the primary publishing language. It runs through a multi-step detection process to reach a confident result.
Detection Pipeline
Every text, audio, and metadata signal from the creator's profile across all platforms.
The Confidence Threshold
Every detection comes with a confidence score. The pipeline only publishes a language when confidence clears a minimum threshold. This prevents a single Italian quote from outweighing fifty English captions.
Where the Minimum Sits
A creator posts mostly in Spanish, with a couple of English motivational quotes and a single short Italian phrase. The detector returns very high confidence for Spanish, low confidence for English, and almost none for Italian. Spanish clears the minimum easily; the others do not. Result: language = Spanish. The Italian phrase is excluded.
Language Detection for Two Illustrative Profiles
- Instagram captions: English
- YouTube descriptions: English
- TikTok captions: English
High volume of content samples across 3 platforms
- YouTube descriptions: English
- Instagram captions: mixed (English + Portuguese phrases)
Fewer total samples, 2 platforms only
More platforms and more content give a stronger language signal. NikolaTravels publishes consistently in English across three platforms, so the algorithm has high volume to work with and returns a very confident result. BorisLifeLog still passes the confidence threshold, but the mixed Portuguese phrases from travel content and fewer total samples produce a lower score. Mixed-language content is common for travel creators, so the algorithm weights by volume of consistent-language posts, not just the presence of any language.
Cross-Platform Signal
When we match a creator's profiles across platforms, language data from one platform can fill gaps on another. If a creator publishes enough content in one language on one platform for a confident detection, but their profile on another platform is too new to carry enough signal, the matched profile carries the language over.
How a Language Fills Across Platforms
Too few posts on TikTok for detection, now filled from matched YouTube profile
Our profile matching system links accounts that belong to the same creator. When a language is detected on one platform but the other lacks enough content for a confident result, the matched value fills the gap. If two platforms detect different languages (for example, English on one platform and Portuguese on another), both languages are kept and no override happens.
Validation 
Every accuracy number on this page comes out of a two-track validation process. We compare at least two independent AI sources on tens of thousands of records, and we manually validate at least 1,000 records per subset, with subsets defined to cover every meaningful use case.
