CHALLENGE In a bid to capture a share of the booming short-form video market and engage a younger demographic, a media giant developed a short-form video app modeled after TikTok from scratch. The app scaled to 6 million monthly active users (MAU); however, user engagement was below the benchmarks of competitors.
SOLUTION AlumniHub developed and introduced a recommendation system that:
Gathers User Data: Collects information on app interactions, such as completion rate of videos watched, likes, comments, searches, etc.
Analyzes Video Content: Examines video attributes like music genre, language, objects, scenes, text captions, and hashtags.
The recommendation engine uses ML algorithms to identify patterns between user data and video content, enabling it to recommend:
Similar Videos: If a user enjoys videos about cat adoptions, they'll likely see more animal rescue content.
Complementary Videos: A user watching makeup tutorials might be recommended skincare routines.
Content from Rising Creators: The system can identify promising creators based on engagement metrics, promoting discovery.
RESULT
20% Increase in Average Time Spent: Achieved a significant boost in user engagement, with a 20% increase in average time spent on the platform.