Here’s how & why CDPs differ from CRMs & DMPs

Even if you don’t have one, you might have heard about Customer Data Platforms (CDPs).

It’s a “marketer-managed system that creates a persistent, unified customer database that is accessible to other systems,” according to an institute set up to promote and research them.

And you may be looking at that Customer Relationship Management (CRM) system you have, or thinking about the Data Management Platform (DMP) you use, and asking yourself, “They each have a lot of data about my customers, prospects and targets. Aren’t they essentially the same?”

Glad you asked. So we pinged Buck Webb, VP of Cloud Solutions at RedPoint Global, for his perspective, since his company offers a Customer Data Platform.

The three systems — CRMs, CDPs and DMPs — each have different characteristics and strengths, he told me.

A CRM is designed to record transactional data, he said, like when User X makes a purchase on your website and becomes a customer. Or when another user calls the contact center with a complaint.

Those can be connected together as being the same user through a “hard key,” like the same phone number or same email, so you know they are the same person.

But CRMs, Webb added, are not designed to do all kinds of identity matching.

“It’s quite common to have duplicates” in CRMs, he said. The companies behind them “are not being disingenuous or deceptive,” he said, since that’s how CRMs have largely operated.

Part of channels

Webb contrasted that to CDPs, which he said “brings a data management layer to the table” that CRMs don’t have.

It uses this layer to link different pieces of customer/prospect info into single identities, he said, either deterministically (a hard key, like a phone number) or probabilistically (a likelihood based on common attributes, like the same physical location of the device at night as detected by IP addresses). It’s not just based on matching through a clear connector.

And “not all customer information is transaction based,” he pointed out.

Customer info can come in streams of data, some of which are purchases or calls to contact centers, while some are visits to websites, activities in an app or demographic information. CDPs are built to accommodate continual flows of information from any source, structured or not, and turn them into single identities.

While a CRM is oriented around transactional data, Webb said, a DMP is built for big batches of audience data that are used to target ads, not for absorbing continual streams. While CRMs and DMPs can both do segmentation, Webb said that CDPs can generate them continually and with more sophistication, such as “finding customers with a lifetime value greater than X,” where X is a continually changing value.

Can a CRM be tuned into becoming a CDP?

Perhaps, he said, but “it’s better to use a CRM and a DMP” as part of channels.

He suggested looking at the CDP “as the central core database” of customers, where all data flows and where the master file of identities is managed and kept.

“When somebody buys something, send out an email via the CRM,” he said, since it is likely already connected to an email campaign tool. It’s also the repository that is integrated with the call center, so keep that connection.

And the DMP, he noted, is part of the ad targeting channel, and is connected to an ad-buying demand-side platform.

“In an ideal world, they’re all synced up,” he suggested.