ContentWise is the recommendation engine built for performance.

Multi-platform recommendation capabilities

ContentWise can recommend any digital media content available, such as live-TV programs, TV channels, On-demand content, movies, music, ringtones. ContentWise works with any type of TV service such as VOD, Linear TV, Web TV and it can deliver recommendations to STBs, Connected TVs, mobile phones, email, web applications and any media device.

Fresh recommendations updated in real-time

ContentWise provides real-time recommendations based on the analysis of both historical and real-time data. This feature enables service providers to adapt their offer according to customers instant feedback. Real-time analysis also more accurately profiles users and his/her current ‘mood’. In addition, service providers can set up ageing filters to customer profiles ratings in order to keep recommendations updated according to viewers habits evolution.

Marketing and promotion campaign management

ContentWise provides marketing department with a simple and effective tool to manage and implement marketing campaigns and promotions. Marketing managers can easily set business rules, push or filter content according to predefined and dynamic campaign goals and target user clusters.  The embedded reporting tool enables marketers to track campaign progress and results.

Built on sophisticated recommendation algorithms

ContentWise recommendation engine blends multiple algorithms to optimize the recommendation results and enable fine-tuning of recommendations based on user experience and marketing objectives.

  • Profile-based analysis

    Recommendations are based on the analysis of individual profiles, i.e. preferences defined according to ratings and viewing habits.

  • Collaborative filtering analysis

    Recommendations are based on “Who watched this, also watched…” mechanisms applied to all user interactions.

  • Related-content analysis.

    Recommendations are based on content similarities, provided by metadata information, such as cast, director, genre, synopsis…

  • Community-based analysis

    Recommendations are based on the analysis of user communities and friends’ preferences.

  • Combined rating strategy

    Recommendations are based on a combination of an analysis of TV watching habits, compared to the whole customer base, and linked to explicit ratings.