How Does a CDP Work?

The Customer Data Platform’s ultimate purpose is to unify any and all desired data types and data sources in order to build a complete picture of their customer and then make those profiles available to your marketing systems of engagement.


For example, connecting the CDP to your POS system provides information regarding a customer’s transactions. Connecting to the eCommerce or web presence can provide a wealth of customer detail such as items viewed, visits, cart actions, favorite departments or items, etc. Integrating survey responses or chats will help round out the “personality” and communication style or preferences for your customer through active and voluntary participation. Integrating in-store movement via beacons or WiFi browsing while shopping provides an even greater level of granularity.

Some of this information is available in other systems today. The problem is that those systems were purpose-built for other functions and provide only some of the functionality needed to build a Single-User-Profile. These include:


Customer Relationship Management software was designed for the B2B world – specifically the sales and account management function. As such, they are good at dealing with structured data only and are not connected to operational (such as POS or eCommerce) systems. In other words, CRM offerings are unable to tie together customer activities, behaviors, and/or intent across different channels which is a critical function of the CDP in building and maintaining a Single-User-Profile.


Master Data Management software is similar to a Customer Data Platform in that both include the function of pulling together disparate data sources to build and inform a profile (identity resolution). However, like CRMs, MDMs also lack the ability to work with semi-structured or unstructured data.

Another challenge with MDM is that these offerings are usually part of a larger integration and, as a result, tend to be owned by the IT department whereas the CDP was designed to be owned and maintained by Marketing thereby avoiding the traditional data challenges of:

  • Free access of data
  • Rigid guidelines and structures imposed by owners within other departments
  • Lack of connections to the delivery tools in the MarTech stack


A Data Management Platform is another platform used to connect and manage data. One of the key differences between it and a CDP is that the data contained in a DMP is de-identified – meaning that the data is collected across a large number of users and none of it is tied to a specific user. The data is anonymized. Even worse, the data is not persistent – it expires after a period of time – usually 90 days.

Obviously, if true personalization is your goal, then a DMP is not the right tool.

Another key difference is that the DMP is focused mainly on digital data used to create segments for online campaigns. The Customer Data Platform, on the other hand, is about unifying customer data from across all channels and then using that data to engage customers and create a personalized customer experience across all channels – Online, In-App, and In-Store.


Enterprise Data Warehouses, as their name implies, act as stores for ALL corporate data. As such, they are just repositories of data, unable to unify or attribute data to individuals (identify resolution). Given their purpose, they tend to be very large, complex, and expensive systems owned by IT which are great at running reports to inform business decisions but are a very poor fit for marketers looking to engage customers in a personalized manner.

And the list goes on.

There are other MarTech products out, all promising to create a Single-User-Profile but, in the end, your needs will dictate the functionality you require.