Treasure Data

Treasure Data (backed by SoftBank) is an incredibly massive, deeply technical leviathan that completely dominates the "Global Enterprise Big Data" sector of the CDP market. While Segment focuses heavily on tracking website clicks, Treasure Data was mathematically engineered to ingest petabytes of terrifyingly complex, disparate offline and online data from massive legacy corporations (like Subaru or AB InBev) and unify it. Its signature feature is "The Massive Schemaless Data Lake." Massive conglomerates have data trapped in 30-year-old on-premise servers, modern SaaS apps, and physical Point-of-Sale (POS) systems. Treasure Data doesn't force the company to clean the data before sending it. It mathematically ingests raw, unstructured data via 170+ enterprise connectors, throwing it into a massive Hadoop/Presto-backed data lake, and then mathematically structures it post-ingestion. It heavily dominates "Deep Predictive AI." Once the massive data lake is built, Treasure Data's machine learning engine goes to work. It runs terrifyingly complex mathematical models across billions of rows of historical purchasing data. It can mathematically predict not only which customer will buy a car in the next 6 months, but exactly which model and color they are most likely to purchase, automatically pushing that data to the dealership CRM.