Composable CDP refers to an architecture where customer profiles are built in a company’s enterprise data warehouse, rather than a separate CDP database.  It implies that the tools used to build and access the profiles are components provided by different vendors, all working directly against the data warehouse files.  This contrasts with a conventional CDP, where a single vendor provides all the tools to build and access the profiles in the CDP database.  Composable CDP advocates argue their approach avoids creating a new copy of customer data, leverages existing data warehouse resources, and gives the company greater control over its data.  While most products to support a composable CDP are provided by vendors who specialize in a particular function, such as reverse ETL, some conventional CDP vendors have unbundled parts of their system to support a warehouse-based approach.

The composable approach is ultimately one method for building a CDP system rather than buying a conventional CDP package.  Other options for building a CDP include building the tools internally or using a mix of built and bought tools.  The actual advantage of building a CDP depends largely on how much change is required for the existing data warehouse to support CDP requirements.  If many new capabilities are needed, it is often more effective to use a conventional CDP instead.  There are also specific capabilities, notably real-time data updates and access, which are poorly supported by most data warehouse technology.  Achieving these may require copying the warehouse data into special structures that effectively create a separate CDP database.

This section provides a training course on making the build vs buy decision, a Composable CDP Self-Assessment tool to identify gaps that must be closed to meet CDP requirements, and links to papers with additional information on composable CDPs and building vs buying a CDP.