Switching Behavior Analysis and Propensity Scoring
Media & EntertainmentBusiness Impacts
Optimized
Ad campaigns
5x
Increase in Conversion Rate
Customer Key Facts
- Location : Mountain View, California
- Size : 201-500 employees
- Industry : Media & Entertainment
Problem Context
The client, a TV data company, wanted to determine the switching pattern of viewers and identify users who have lapsed a particular show. They also wanted to assign a propensity score to each user having a higher tendency of watching a new show on TV to segment the group and target them efficiently.
Challenges
- Inconsistent attribute details in the metadata
- Multiple sources of data resulting in disparity in granularity
Technologies
Amazon S3
Amazon SageMaker
Amazon Lambda
Amazon Athena
Leveraging Statistical Analysis and ML-Driven Clustering for Switching Behavior Analysis and Propensity Scoring
Solution
Quantiphi leveraged Statistical Analysis and ML-driven clustering to generate insightful viewership patterns, understand switching behavior and viewership preferences. It helped the firm target the right set of audiences to optimize and improve the Ad campaign.
Result
Generation of Insightful viewership preferences in real-time leading to better campaign optimization
