Don’t conflate CDP and CRM

Responding to my editorial in this morning’s Daily Brief, “I say CDP, you say CRM,” David Raab, founder of the CDP Institute, took me to task for perpetuating some confusion. With his permission, we publish his comments here.

There’s a significant difference between a CRM and a CDP, and you do your readers a huge disservice by conflating CDP and CRM.  Yes, both store customer data – as do data lakes, data warehouses, marketing automation, email engines, personalization tools, web content managers, and a host of other systems.  Each of those is designed for a specific purpose and stores customer data in a way that fits that purpose. 

CRM also has its own purpose – to support sales and service agents when speaking with customers – and is optimized for it.  CRMs are notoriously bad at dealing with data that was imported from elsewhere, and with unstructured and semi-structured data types.   They’re generally poor at sharing their data with other systems.   

CDP has a different purpose – to combine all customer data into sharable profiles.  It’s designed for that purpose, which means it has connections to hundreds of sources, uses a data store that handles all data types, employs advanced identity matching methods, and gives other systems easy access to its data. 

Many CDPs do more, with analytical, campaign, and delivery functions, but those are not what makes them a CDP.  As it happens, very few CDP vendors also extend to CRM functionality, precisely because the technical requirements are so very different.

This confusion is not a new problem. 

Several years ago, CDPI Institute launched its RealCDP certification program, which sets seven criteria to be considered a true CDP: accept data from all sources, retain all detail of the original input, store data as long as the user wants, construct unified profiles, share profiles with other systems, react to events in real-time, and share individual profiles in real-time.  We chose those because we feel they are capabilities needed to support the use cases that people expect a CDP to support, such as giving CRM users a view of interactions in other channels, updating retargeting lists in real-time based on ecommerce transactions, and combining in-store and on-line transactions to build campaign lists. 

The common thread is combining data from different systems and sharing the result across all systems.  This is what CDPs are uniquely good at.  The fact that every major CRM vendor has developed a separate CDP product – despite considerable initial resistance to the concept – is the most compelling evidence that they are truly different systems.