According to Gartner and their recent survey, 50% of enterprises are equipped with a DMP (Data Management Platform).
The promises in investing into these DMP or Big Data solutions is huge. From Consumer Product Companies (CPG) to retailers, e-commerce and media groups, most of them spent a few thousands of dollars in the last two to three years to access this promise. But it’s not as simple to make the data collected actionable.
Moreover, the recent notion of CDP (Customer Data Platform) is bringing even more confusion for CMO’s (Chief Marketing Officer) to really understand where the frontier stops between a DMP and a CDP. This post aim is an attempt to clarify it.
What is a DMP (Data Management Platform)?
“DMPs gather data, organize, configure it and make it available to various users and systems” according to Gartner. For marketers, it is a way to improve the ad targeting, personalization, channel integration and audience retargeting. We recognize that some DMPs have added CDP features but it’s key to understand that it’s not what they were originally designed to do.
What is a CDP (Customer Data Platform)?
“CDPs ingests customer-related data from multiple sources systems, cleans and unifies it to create a single customer view and makes it available to other marketing systems to use. Some CDPs provide additional function such as analytics, predictive modeling, content recommendations, and campaign management.” (according to David Raab from the Customer Data Platform Institute).
What are the essential differences between DMP and CDP?
- DMP cookies are generally anonymous (OK, it might be enriched by 3rd party data providers) whereas CDP records are linked to identified individuals.
- DMP data is a list of attributes while CDPs store very detailed information such as purchase transactions, CRM, web / mobile behavior and offline data like PoS.
- DMP is mostly a data warehouse or data lake when a CDP (out of the data layer) has built-in two additional levels called a “Decision Engine” and “Delivery Integration” to make the data actionable building user journey and connecting in plug & play to existing marketing system.
- DMP does not do the cross-channel identity resolution needed to build a single customer view whereas CDP are built for this purpose and includes bulk processing and extracts.
- DMP are enterprise-wide projects very specific to each organization while CDP proposes a more flexible format with input and output connectors for both data and interaction systems.
- DMP will take 10 to 18 months for a full setup for a cost above $100,000 whereas CDP can be implemented in 2-4 weeks for a monthly recurring fee between 3 to 12,000 $ per month. It will show ROI in a few months using A/B testing.
- DMP will require a strong involvement from the vendor or a service partner to handle all the technical work before they’re ready to work whereas CDP will provide both connectors and interfaces that make it easy for marketers minimizing the organizational issues.