Personalized OTT platform

Personalized OTT platform

Challenge

Client is an OTT service operator launched by one of the biggest  telco and mobile operator in a CIS country. They decided to measure the effectiveness of their current personalization tools, skills and capabilities provided by a local vendor. Innovation board selected TV domain for a pilot to compare external solutions to the local vendor in live environment.

Approach

Both local vendor and us implemented and operated three Content Recommendation placements for the client’s TV domain, using data collected on Android devices. The recommendation boxes were placed on the Main Page “Best in Subscription”, “Recommendations” on Recommendations page and “Similar films” or “Similar tv-series” on film/series page.

Solution

The machine learning algorithms were optimized for conversion to sales, as this was selected KPI by the client. From six placements (3 scenarios for Android and 3 for non-Android) there was one, where the two solutions delivered the same result. In the rest of the 5 other placements our recommendations resulted in up to 58% higher conversion to sales, and finally 17% higher revenue versus the other vendor.

Information

  • ClientTelecom Company
  • CategoryPayTV
  • Tags
GET STARTED

We are happy to open up "the black-box machine learning modeling" for your company