A wooden boardwalk in dense forest that splits to the left and right.

Which Customer Data Platform Road Should You Travel?

7 minute read
Dom Nicastro avatar
Customer data platforms (CDP) are more than a one-flavor piece of software. We've got some tips for finding the right CDP for your organization.

Nothing seems to slow down growth of Customer Data Platforms (CDPs). Not even 2020.

The industry added 13 vendors, 800 employees and $250 million in funding over the last six months of 2020, according to the Customer Data Platform Industry Update: January 2021 by the CDP Institute. The CDP Institute estimates industry revenue will reach $1.55 billion in 2021, a 20% increase over 2020.

“It's clear now that digital transformation is here to stay. It’s truly going to happen and CDP is absolutely the platform to do that on top of,” David Raab, founder of the CDP Institute, told CMSWire. “The CDP industry held up quite well during the pandemic, and we really didn't know if it would.”

So now the bigger question is what kind of CDP is right for you if you’re going down that path to manage and cultivate your customer and prospect data. Because all CDPs are not created equal, it’s worth exploring how CDPs are defined and for which segments CDP vendors fit.

CDP Defined

A CDP is a data management system that has a unified and persistent database, which can be accessed by other marketing technology (martech) systems, according to CMSWire's CDP Buyer’s Guide published in 2020. CDPs ingest and integrate behavioral, transactional, structured and unstructured data from multiple sources into a single repository that allows a business to build a unified profile around an individual customer.

Types of CDPs

According to the CDP Buyer’s Guide, CDPs come in four different flavors:

  • Data: These CDPs ingest customer data from multiple sources, combining and linking that data to customer profiles. The resulting data and profiles are stored in a persistent database that is available to external martech systems. Platforms in this category will typically have specialized features for data management and unification. Many CDPs in this category started as tag management or web analytics systems.
  • Analytics: These platforms have the data management functionality of data CDPs with expanded analytical and visualization features. Analytics CDPs may include the ability to do customer segmentation and can have advanced features like machine learning and predictive modeling. These systems can sometimes automate the distribution of segmented lists to marketing automation platforms.
  • Campaign: Campaign CDPs are more inclusive systems that provide data management, analytics and customer treatments. Treatments differ from segments in that they may include personalized messaging, outbound marketing campaigns, real-time interactions, and product and content recommendations. These systems may include the ability to orchestrate customer treatments across multiple channels.
  • Delivery: These platforms have data assembly, analytics, customer treatments, along with message delivery, typically through email, web, or CRM. These systems often started as delivery products and added in CDP features later.

Campaign and Delivery vendors now account for more than 70% of companies and employment, according to the CDP Institute. Many of these are older firms that added CDP functions to existing products, researchers noted.

“Originally, it was much more focused on the data and analytics CDPs, so that's definitely been a change over time,” Raab told CMSWire. “I expect that will grow even further. … We just added Delivery CDPs as a category fairly recently and that's been growing within that group, kind of taking away from the campaign CDPs.”

Related Article: What's the Difference Between a CRM and a CDP? And Why You Should Care

Will Your CDP Provide Better Data-Driven Activation?

Finding the right CDP fit starts with knowing how — and if — your CDP will fit into your marketing technology equation.

“Choosing the right CDP is like making a long-term investment and can be challenging, as many of them provide excellent services for brands looking to invest in/maximize their data strategy,” said Oz Etzioni, CEO of Clinch. “One thing to consider is the CDP’s ability to integrate with a variety of tech vendors who could provide even more value now, or down the line, in building a better data-driven activation, faster and more granular audience segmentation, and a more holistic view of your consumer. The right CDP will be able to help you evaluate and identify areas in which you might be leaving valuable insights and opportunities on the table.”

In-House, Service Partner Capabilities

Identifying a best-fit CDP is contingent upon an understanding of the organization’s current and future capabilities, both in-house and through service partners, according to David Markel, vice president of enterprise strategy at Publicis Sapient. The functionality that comes with any CDP, bare-bones or fully-featured, should be considered in light of the presence or absence of resources, either human or technological, that provide for some or all of this functionality for the organization today, he added.

“It comes down to the additional value any type of CDP can provide on top of an organization’s existing capabilities,” Markel said. “A CDP focused on the core task of creating unified customer profiles could fill a capabilities gap around data engineering, devops, etc. For organizations that have the requisite data science, business intelligence, and campaign activation resources required to deliver high-value use cases leveraging a consolidated customer data asset, but lack the engineering capability to create this asset in the first place, this type of CDP could be a great solution.”

Fully-featured CDPs, meanwhile can supplement or enhance the work of teams that are downstream of the consolidated customer data asset, Markel added. “For organizations where data science talent is already over-allocated, the turnkey modeling and ML functionality offered by fully-featured CDPs can increase the impact of an organization’s customer segmentation beyond a rules-based approach,” Markel said, “and can improve the campaign performance of different end users across marketing, sales, supply chain, etc.”

Similarly, the robust data visualization features offered by this type of CDP could supplement an organization’s existing BI capabilities, Markel added, and offer another experience for end users to obtain their preferred visualizations in addition to an organization’s existing tooling.

Learning Opportunities

Related Article: 3 Predictions for the CDP Market in 2021

Reading Dynamic Data Signals

The CDP you select needs to operate as a source of truth for understanding historical trends as well as reading dynamic data signals and producing instantaneous customer profiles, according to David Hegarty, senior director of solutions for R2integrated. “These profiles will constantly change based on what’s happening online and offline,” Hegarty added.

“It’s imperative all the activity is gathered in real time; otherwise, you could be looking at data from 30 days ago and lose the context of where customers fall in the purchase lifecycle.”

Since providing exceptional customer experiences is a focal point for marketers, you must select a CDP that leverages data to orchestrate experiences based on behaviors of audience segments and profiles, Hegarty added.

“The leading CDPs will integrate with other marketing solutions to generate memorable customer experiences like enticing offers in emails or SMS messages with discounts,” Hegarty said. “Understanding the data sets, and how you wish to use it to engage will assist an organization in defining their CDP requirement.”

Creates, Manages ML Models

Does your CDP give you the ability to create and manage machine learning models? This puts the power in the marketing team’s hands, Hegarty said, to democratize AI and build analytics on the go instead of waiting on help from IT. “By quickly acting on these models,” he said, “you don’t have to worry about responding too late to customer needs.”

The best CDPs also include simple user interfaces for people who don’t have a technical background. Building a machine learning model is a much more streamlined and frictionless process this way, according to Hegarty. “After you put a model in place, it’ll automatically kick off experiences based on customers’ behavior and data — like having a 24/7 marketing manager on your team,” he said.

Related Article: Beyond CDP or Bust: The Looming CDP Reality

Multiple Marketing Use Cases

Well-designed CDPs provide value across many marketing use-cases, Hegarty said. For example, if you’re focused on customer retention, you can analyze holistic customer profiles to gather insights that are most valuable to your customers and plan an effective loyalty program strategy.

“You can deliver consistent messaging throughout email, push notifications, online encounters and paid advertising that’s also relevant to the messages your customers want to receive within personalized campaigns,” Hegarty said. “This payoff is huge and allows you to connect with customers during essential purchase moments to give you the competitive edge.”