Since not all influencers properly disclose whether a post is sponsored, and only a few countries have a disclosure policy, HypeAuditor uses a machine learning-based approach that can find out if a certain post is sponsored or not. Such posts are identified as Likely Sponsored.
Likely Sponsored posts model:
During analysis the model considers different factors to determine the type of each mention in the post:
Number of post mentions
If the mention is a brand
If the author is a brand
Category of the mention
Category of the author
Type of post: video, image, reels, etc.
Number of likes, comments
Account size of the author and the mention
Author's and mention's business category
If mention has a URL in the profile
Mention place in the text (if a mention comes first among other mentions, it is more likely to be sponsored)
Additionally: if there is a sponsored hashtag (like #promo, #ad) and the post has only one mention, that mention is considered sponsored. If there is a sponsored hashtag and several mentions, all mentions are considered Likely Sponsored.
The current accuracy rate is 95%. However, this accuracy is always changing when the machine learning model is updated. It is not a static algorithm — HypeAuditor regularly trains the system with paid posts so that it can learn how to classify new content.
By surfacing probable sponsorships through AI, the Competitor Analysis Report provides a more accurate and complete view of how brands invest in influencer marketing — enabling better benchmarking and budget planning.
To learn about how undoubtedly paid content is detected, please check the article about Sponsored posts.