11% match rate improvement with adaptive identity graphs
Created an adaptive identity graph that automatically maps or un-maps an identity to a person in near real-time, maximizing customer acquisition and retention across markets.
The client is a leading IT company and wanted to build an Identity Graph, which can be utilized to match identities and highlight possible signals of intent for the acquisition and retention of consumers. The current computation time for creating graphs was too high, coupled with a lack of a well-defined testing framework to understand the quality of the identity graph and benchmark improvements.
Sigmoid built an adaptive and self-learning identity graph that will automatically map or un-map an identity to a person in near real-time. We enabled deep drill down on data for understanding the time-based behavior patterns to identify bad behaviors of lead creation.
Our adaptive and self-learning graph solution led to a 15% improvement in predictive power and an 11% improvement in match rate.