AI Powered Customer Analytics in the age of GDPR

Online and offline customer interactions generate prodigious amount of data that can be analyzed by AI and ML algorithms in order to improve customer experience. The recently enacted European General Data Protection Rules (GDPR) grants customers control over their data including the right to be forgotten. The big data technologies that are capable of analyzing massive customer interaction data (e.g., HDFS, Kafka, Spark, Flink, and Beam) are most performant when working with immutable data, which does not bode well with the right to be forgotten. In this presentation, we briefly discuss the implications of GDPR on various behavioral data generated by websites, apps, in-store, and ad impression data. We show how GDPR requirements can be efficiently satisfied using an architecture that isolates a GDPR-safe analytics zone, which is the only area accessible to the users, from the input and output zones where encryption, decryption, and key-management is tightly controlled to satisfy the GDPR requirements.