Data Management Platforms (DMPs) came on the scene a few years ago and are seen by many as the savior of marketing. Here, they say, we can have the “golden record” for our customers. In the DMP, vendors promise that you can collect all the information you need for a 360-degree view of the customer.
The only problem — it’s just not true.
Gartner defines a DMP as
Software that ingests data from multiple sources (such as internal CRM systems and external vendors) and makes it available to marketers to build segments and targets.
It happens that a number of DMP vendors make up the core of Gartner’s Magic Quadrant for Digital Marketing Hubs (DMH). Gartner analysts anticipate over the next five years the DMP will turn into a DMH, providing:
Marketers and applications with standardized access to audience profile data, content, workflow elements, messaging and common analytics functions for orchestrating and optimizing multichannel campaigns, conversations, experiences and data collection across online and offline channels, both manually and programmatically.
But DMPs were originally designed around one channel: online ad networks. When DMPs first arrived on the market, they helped websites deliver the best offers by utilizing cookies to track a person’s web activity anonymously. They then morphed into adtech as part of a programmatic buying process, essentially helping companies market to a specific kind of segment. They’re great for this single purpose, but begin to fail when they’re asked to do more multi-channel campaigns that utilize machine learning for a more targeted approach.
Because data stored within a DMP is anonymous, the DMP can be helpful for segmented online advertising. It doesn’t necessarily need to know who you are to serve up an online ad based on your previous web surfing history. While it’s true that marketers can link plenty of first-, second- and third-party data to cookies housed in a DMP, it’s basically just a data warehouse and nothing more. DMPs cannot store as much data as a relational or Hadoop-based system.
Most importantly, you cannot use DMPs to store any personally identifiable information (PII) – the molecules that help create the unique DNA for each of your customers. As a marketer, if you’re looking to take all of your first-, second- and third-party data to create a system of record for your customer, then a DMP simply won’t cut it.
As we future-proof our technology investments in the age of the Internet of Things (IoT), a DMP can’t compare to a Customer Data Platform (CDP) for achieving that elusive “golden record.” CDPs do something unique – they can capture, integrate and manage all types of customer data to help create a complete picture (including DMP behavior data). However, to what degree and how this is achieved varies widely from vendor to vendor.
CDPs were designed from the ground up to capture, integrate and manage all types of dynamic customer data, including data from social media streams and the IoT. To that end, they’re based on relational or Hadoop-based systems, making them better able to handle the deluge of data that lies ahead as more IoT-oriented products come online.
This is why Scott Brinker separates DMPs and CDPs in his Marketing Technology Landscape Supergraphic. Called out in his squint-inducing 3,900+ logo chart are two separate categories with different vendors.
In his write up announcing the graphic, Brinker points out correctly that the One Platform to Rule Them All idea has never truly come to fruition, and what exists instead is a cobbling together of platforms to perform certain tasks. Marketers turn to one solution for email, another for web, another for data and so on.
What marketers need is not a large platform that does it all, but a data platform that gives them the information they need to make decisions.
The truth is, that both Brinker and Gartner touch on something that is just starting to emerge: a true orchestration platform. Built on CDPs, these are designed for true omnichannel marketing, giving marketers the tools they need to make and execute data-driven decisions across all channels.
As marketers prepare for tomorrow, they are going to need to make buying decisions about their data platforms today that will affect how they’re used in the future. Choose wisely and you will have a platform that will help bring everything together. Choose poorly and you’ll be back at square one in a short amount of time.
The State of AI in Marketing: Activating Customer Data for AI Powered Marketing (from aspirations to reality)
The first ever State of AI in Marketing Report of its kind dives deep into how leading brands are using AI and making the most of their Customer Data
You must log in to post a comment.