Case Study: Sharethrough achieves faster prototyping and improved engineering productivity with Databricks

Sharethrough - Customer Case Study

Sharethrough is a native advertising platform that processes terabytes of clickstream data from high-traffic publishers (for example, Forbes and People) to measure product effectiveness and inform product and support teams. Their initial self-hosted Hadoop + Hive setup required two full-time engineers to manage infrastructure and suffered from slow, unreliable ad-hoc queries, which severely hampered exploration and team productivity.

Deploying Databricks in an AWS VPC gave Sharethrough fully managed Spark clusters and an interactive notebook workspace within days, enabling fast, memory‑optimized queries, real-time visualization, and multi-language development. The move accelerated prototyping, made debugging complex pipelines easier, freed the two engineers to focus on core work, reduced downtime, and materially improved overall engineering productivity.

Sharethrough

Robert Slifka

Vice President of Engineering