Data historians and time series databases are often treated as competing choices for industrial data stacks, but they serve fundamentally different purposes. Historians excel at OT data collection from plant-floor sources (OPC-UA, Modbus, SCADA) with certified data integrity, while time series databases provide the analytics layer for cross-asset queries, OEE calculations, and BI tool integration. Replacing a historian with a time series database means rebuilding OT connectivity and re-certifying data integrity — a costly project that doesn't solve the actual analytics problem. The recommended architecture keeps the historian for ingestion and adds a time series database like CrateDB as the analytics layer, enabling real-time cross-plant SQL queries without disrupting existing OT infrastructure.