Case Study: Comcast delivers personalized, scalable smart homes with Neo4j

Graph Technology Helps Xfinity Create Personalized, Smart Homes

Xfinity from Comcast, which serves over 30 million U.S. customers, faced a common smart‑home problem: connected devices could report events but lacked the semantic and social context needed for personalization and automation. Without a way to model relationships between people, locations and devices, homes couldn’t perform person‑centric commands (for example, “turn off the lights in Lily’s room”) or deliver tailored experiences.

Comcast built the Xfinity profile graph — a household‑level, multi‑tenant user‑profile service implemented on a native graph database and exposed via GraphQL on Neo4j — to encode rich definitions and relationships for people, places and devices. The graph enables per‑household and per‑person controls, faster feature delivery (such as personalized xFi Wi‑Fi profiles and guest/pausing device controls), and a scalable, flexible platform that can easily extend to new data types and products.

Comcast

Jessica Lowing

Director of Project Management