Customer Data Platforms 101 [Guide]
Customer data platforms have become one of the most popular technologies in the marketers' toolbox in the past few years, but there is still a lot of confusion around what they are, how they work, and their capabilities.
No topic has generated more interest and debate among the marketing technology community over the past few years than the “customer data platform” (CDP), the category of software designed to help marketers fuel one-to-one engagements, and enable unified insights, in a cross-channel world.
Detractors have tried to position CDPs as a fresh coat of paint on old technologies like the data management platform (DMP), tag manager, or any number of CRM applications. Not helping matters, several point-solution vendors finding themselves in need of a new market opportunity have embraced the term CDP. Still, the allure of CDPs is that they represent nothing less than what one analyst refers to as “a generational and procedural shakeup of enterprise marketing technology” and, if only for that reason alone, need to be reckoned with by everyone.
This guide will help you understand
- What customer data platforms actually are
- The business and operational drivers underpinning customer data platform adoption
- The different types of customer data platforms
- How to determine which customer data platform best suits your needs
What is a Customer Data Platform?
The CDP Institute’s standard litmus test for whether a data platform qualifies as a CDP is three-pronged. A true customer data platform is marketer-managed; serves as a persistent, unified customer database; and it must be accessible to other systems. Sometimes it’s easier to understand CDPs by what they are not.
- Unlike data lakes, data warehouses and most CRM systems, which are controlled by IT, marketing departments are the primary stakeholders and system owners.
- Unlike web DMPs, CDPs capture and store personal identifiers and persist data over long periods of time.
- Unlike big marketing clouds, which bundle data management and execution, CDPs specialize in unifying data and delivering it to other systems for execution.
Three prongs of a CDP
Engineering debt has always been a problem for marketers looking to do advanced customer data querying and analysis. As much as IT may want to help, constraints on their time mean they have to prioritize core product development as much as possible, leading to marketing request delays. A CDP changes this by eliminating marketers’ need to ask engineering in the first place; instead, once a CDP is implemented properly, marketers can easily update customer data, create and forward segments, A/B test content across channels, and more with just a few clicks. By taking out the engineering middleman, marketers are free to test and change campaigns on the fl y to minimize wasted spend and maximize campaign ROI.
Persistent, unified customer database
A critical part of marketing is knowing who you are marketing towards, but the proliferation of channels and devices have made knowing who your customers are more difficult. A customer might first engage with your brand on their laptop at work, then continue browsing on their phone during their commute, and finally make a purchase on their tablet while at home. Without the ability to connect all of these data points to a single customer profile, marketers may see them as three separate people and waste money targeting towards a customer that has already made a purchase, or worse, annoy them enough to lose their future patronage.
A customer data platform is able to take all of these separate data points and attributes and match them to a single view of your customer, so marketers can avoid targeting customers with the wrong message or offer. The main difference with this kind of profile versus customer profiles developed through other systems is its persistence. Instead of relying on cookies, which collect and anonymize data and have a set expiration, CDPs collect first-party data like mobile identifiers, geographic location, emails, and more for a persistent, robust customer profile.
Accessible to other systems
Simply having a unified customer database isn’t enough; you need to be able to use that database effectively. A CDP should not only collect and match customer data from disparate systems, it should also be able to forward it to the services and systems you’re using to execute campaigns. For example, a data point collected as a result of a customer browsing for boots is collected, matched to their customer profile, then this profile is used to activate a push, email, or display ad for boots. This connection, although simple-sounding, is critical for marketers looking to increase the speed, scale, and accuracy of their targeting. By cleaning and matching the data, a CDP makes customer data actionable and provides the connection to the systems already in place to make use of that data.
Core features of a customer data platform
Combining profiles and associated data points from multiple platforms and systems into a single customer record.
Creating segments and syncing those segments to diverse execution channels
Streaming data to various analytics systems (preferably in real time) to enable modeling and insights
Monitoring inflow and outflow of data, reporting variances or anomalies, and maintaining data schema
Combining profiles and sending them out for insight and executional purposes. Then appending modeled and exhaust data back from analytics and engagement systems. This creates a complete, well-rounded understanding of customer behaviors at a granular level.
Taken together, these represent the foundations for the modern, identity-driven marketing technology stack.
While the concept of customer data platforms has been around for a long time, the recent surge in interest is owing to the confluence of what might be labelled as “demand” and “supply” side factors:
- Demand Side: The customer journey has become more fragmented: the average consumer has at least four connected devices with that number expected to increase to more than six by 2020. Mobile, of course, is a big part of this. Customers increasingly treat mobile as the hub for their online and offline brand experiences and are turn to their connected devices in their moments of need—whether that’s at home, in-store, or on the go. In order to meet the expectations of the nearly half of Americans who have made the “mobile mind shift,” brands must anticipate and address “in the moment” needs, connecting channels and data sets at a speed and scale like never before. This requires a holistic, modern data infrastructure.
- Supply Side: Marketers have accepted no single “cloud” will solve all their problems: The Big Five providers—Google, Adobe, Salesforce, Oracle and IBM—currently account for only a minority of technologies in the average large company’s marketing stack (the big clouds have relatively little footprint in the mobile ecosystem, for example). What’s more, most of the these companies use more than one Big Five product. In other words, no matter which cloud platforms is used, businesses still need a multi-vendor strategy to create a single customer view. CDPs provide the “connective tissue” betwixt and between all these different cloud software providers.
The types of CDPs
There are many different types of CDPs. Here are some of the factors to consider when trying to determine which type of CDP is right for your business:
- Geography: CDPs vendors differ a great deal in their privacy, security, and governance standards. For example, many mid-market-oriented CDPs do not offer partitioned data storage, role-based permissioning, or controls for EU privacy compliance, which are vital for companies with a global scope or ambition.
- Partner ecosystem: Different CDP vendors may offer different types of integrations. Prospective CDP buyers should ensure that the CDPs they are considering support platforms and tools that matter most to their organization, and that these integrations are well-documented and certified. See these examples of certification pages from Adobe and Oracle.
- Data collection: Many CDPs are rooted in an email or web-centric paradigm, meaning they aren’t well-suited for mobile and omnichannel marketers. If your customer journey includes people interacting via mobile platforms (Android and iOS), or connected devices (AppleTV and Roku), you’ll need a CDP that supports these natively with robust and well-documented SDKs.
- Data enrichment: By definition, CDPs enable unification and distribution of “first-party” customer data. However, if you wish to learn more about your customers from other sources of data, you’ll need a CDP that facilitates data enrichment from partners and third-parties; and
Data activation: Not all CDPs support the same set of marketing scenarios. Some are great at feeding data into data warehouses and reports but require manual processes to take action on the data. Stills others provide exeutional support but only across a limited number of channels, such as email and sales automation systems. mParticle’s CDP was the first to support the newer breed of mobile marketing automation and attribution platforms, in additional web DSPs, DMPs and more traditional CRM and campaign management applications. This makes it possible to create and orchestrate experiences across the full breadth of the customer journeys.
The ROI of a CDP
Just as there many types of CDPs, there are also many different business drivers leading companies to adopt them.
- Maximize marketing and advertising investment returns: As digital marketing matures and more companies invest in it, many brands have started to see diminishing returns. A CDP can help overcome “tactic fatigue” and take programs to the next level, realizing the full potential of email marketing, paid media and other addressable channels. Customer data platforms feed personalization algorithms with the data they need to reach their global maximums -- Gartner has estimated that companies who employ personalization will outsell by 20% companies that do not;
- Meet and exceed customer expectations: According to Wunderman, 87% of people say they measure their experiences with a brand - any brand - against the likes of Netflix, Uber, and Amazon. CDPs make it possible for companies to create the seamless, highly personalized experiences people expect, despite all the organizational and technological complexity that may lie beneath;
- Create more accountability: By seamlessly integrating customer data across all analytics, measurement, media channels and marketing automation tools, CDPs make the dream of full-lifecycle customer journey analytics a reality, in turn allowing companies to hold marketing more accountable for product investment decisions;
- Improve IT efficiency: Collecting and managing data in silos involves maintaining more data, in more places, than necessary. Forrester Research finds 48% of marketing cloud customers say they have redundant systems and 41% describing their technology portfolio as “too complex.” While most systems can theoretically be integrated with enough time and money, the ideal CDP greatly reduces workload and creates flexibility when services must be replaced;
- Mitigate privacy and security risk: Letting data and tools connected by poorly maintained point-to-point integrations “run wild” increases the likelihood of data leakage. Given the sensitivities associated with customer data, it’s critical to have a properly governed central data layer. The right CDP can provide an extra level of security and control; and
Reduce wasted media spend: B2C companies can save millions of dollars by using CDPs to ensure ad are reaching theright person, at the right time. As a large portion of global media spend moves to addressable channels, this can be done by suppressing audiences on Facebook, Google, Snapchat, Pinterest, and other platforms not interested in your product.
What's next for customer data platforms?
As more data is created by customers during their buying journeys across a growing number of devices, legacy data platforms and integration methods will no longer suffice. Every brand will need a customer data platform purpose-built to integrate and orchestrate experiences in an omnichannel world.
The demand for high-quality, connected data will further increase as machine learning is commercialized and deployed throughout the enterprise. This is because AI can only “learn” through access to large, well-structured data repositories (sloppy data lakes won’t cut it). CDPs will provide the perfect training ground for marketing, sales, and customer service AI.
Finally, CDPs will need to stay ahead of shifting privacy regulations and increasing security threats. It is essential that they innovate, not only along the dimension of customer experience but also for security, reliability, and identity management.
Marketers are no longer dealing with a finite number of systems but rather an ever-increasing number of SaaS applications that house and action data. Meanwhile, the number of places and ways customers are creating data has also increased, from one or two devices per user, to multiple, interconnected devices. Companies need a CDP that can help them overcome the customer data silos, while still allowing marketers to take full advantage of the specialized, best-in-breed components of their marketing stack.
CDPs can enable a wide variety of use cases and benefits for different marketers. With the right data strategy, a CDP acts as a bridge between traditional marketing databases and newer digital channels, between systems of insight and systems of engagement, and between online and offline, to provide a 360-degree view of customers.
As the customer journey continues to fragment, finding the right CDP will only become more important. If you’d like to learn how mParticle can help you unify your customer data, boost engagement, and increase advertising and marketing ROI, contact us today!
Latest from mParticle
GOAT: Lifecycle marketing for scalable growth
Learn how GOAT uses mParticle to streamline their data pipeline and increase Customer Lifetime Value.
mParticle launches new features to help brands create ‘data flywheel’
New features for seamless data quality management, and transformation to serve as a foundation for improved customer experience and better insights.
Better data, better insights, better results: Helping brands create a data flywheel
Introduce total quality management and enforcement into your customer data pipeline with new Data Master features and Calculated Attributes. With Data Master and Calculated Attributes, establishing a source of reliable customer data that will create a customer data flywheel, where the data quality and data’s impact on the product cycle will continuously improve over time.