Customer data platform: What it is and why it’s important

A customer data platform (CDP) is formally defined by the CDP Institute as “packaged software that creates a persistent, unified customer database that is accessible to other systems.” 

Customer data platform: How it works

The basic CDP has four primary capabilities: 

1. Data ingestion. CDPs pull in first-party customer data from multiple sources (e.g., transactional systems, web behavior, call center, demographics or POS). 
   
   
2. Identity management. CDPs resolve the identities of customers across multiple channels. Most use deterministic matching and profile stitching with identifiers provided by the inbound data. The CDP creates and maintains a persistent customer identifier. 
   
   
3. Segmentation. CDPs allow marketers to build and maintain universal, omnichannel audience segmentation with the unified data in the CDP. 
   
   
4. Data provision/activation. CDPs facilitate the activation of the insights and unified customer profile generated in the CDP. This is done by providing connectors and APIs to other marketing technologies. 

Customer data platform: What you need to know

The recent rise in popularity for CDPs has led to a considerable amount of market confusion. Vendors offering CDP solutions come from areas as varied as tag management, digital monitoring products, campaign management, web analytics and data integration. Other groups, although smaller, bill themselves as “pure-play” CDPs. They promote standalone customer database capabilities more like a supercharged master data management solution. Many of these offerings are dissimilar in capability, making comparative evaluations difficult. 

The problems that have drawn marketers to the CDP are very real, and the upside for solving those problems is significant. But there is a growing belief among marketing technology analysts that these capabilities do not meet enough of a marketer’s objectives for the CDP to continue to be a standalone technology solution. Instead, many believe that CDPs will simply become standard parts of large enterprise marketing solution suites, such as multichannel marketing hubs or real-time interaction engines. It’s imperative for companies to carefully consider their CDP options. Marketers need to understand what the CDP solution under evaluation really offers and determine how many of their specific needs will be met for today and beyond. 

Established data infrastructures.
Importing all customer data into a CDP for marketing activity can have significant implementation and data synchronization costs for organizations with established data infrastructure and architecture. We call this the “lift and shift” problem. To work around this issue, companies may want to explore solutions that provide basic CDP capabilities (identity management, segmentation and data provision) without requiring a physical data move into a single database. While these solutions do exist, exploring them may require investigation beyond pure-play CDPs. 

Analytically driven data activation.
While CDPs allow for audience segmentation, the core capabilities typically support rule-based approaches. When algorithmic applications of segmentation enter a brand’s required use cases, real- time decisioning, triggering and next-best -offer execution frequently fall apart. While some CDP vendors are developing analytical capabilities to supplement their core CDP functions, the sophistication of these varies widely. If requirements extend past core CDP capabilities, it may be better to look at purpose-built tools like enterprise marketing suites. These suites offer basic CDP functions but are designed for more broad-based journey orchestration and analytics activity. ​ 

Real-time identity management.
Keeping track of customer behavior across all devices and digital channels and integrating that behavior with offline data in real time is outside the scope of many CDP solutions. If the need includes capturing online events in real time and dynamically updating profiles and offline customer segments, you should look very closely at the CDP to ensure that it can do this. 

Beyond marketing.
A core characteristic of the CDP is providing unified customer data to other marketing applications. But a common complaint is that the integrations to other systems are more complex and time-consuming than advertised. This problem is magnified when customer experience programs extend beyond marketing and into other areas that affect customers (sales, service, fraud and risk). All these areas need to apply analytically driven contextual personalization to their customer activities, and the CDP is the logical place to get the needed customer data. In this situation, the need for real-time event detection, analytically driven decisions and cross-device identity matching will carry over into areas outside marketing. 

Going beyond the CDP with seamless integration

Figure 2 (above) illustrates the seamless integration that will take you beyond your customer data platform. Let’s look at each piece. 

Hybrid marketing architecture: Beyond rudimentary identity management
The variety and volume of data sources contributing to customer profiles require a robust identity management strategy. Capabilities should allow for collection of detailed, contextualized customer-level information about online user behavior (pages, screens and field interactions) while also stitching together online and offline information (CRM, demographics, historical behavior, call center interactions, etc.) into a single unified profile. One that updates dynamically in real time as user behavior changes.  

Real-time digital event detection: Beyond simple data ingestion
Because consumers are using multiple digital devices and channels – many of which require an immediate response – you need to capture events as they happen. Acting on these events in real time (decisioning, triggering, next-best-offer execution, etc.) is also critical. Coupling the need for immediacy with the increasing focus on data privacy will require customizable events to collect all the required data in real time. All while identifying important personally identifiable information (PII), which should not be captured. The solution should provide strong digital guardianship with APIs for adding, merging or removing offline customer information and the ability to encrypt sensitive data in the cloud.  

AI-powered journey orchestration: Beyond segmentation
The real return on investment for marketers will come from activating unified customer data via AI-powered, right-time journey orchestration across all channels. Comprehensive analytics, next-best-offer capabilities, real-time decisioning and seamless integration with other critical marketing functions (planning, testing and attribution analysis) will enhance insights and shorten the time from insight to action.  

Extended customer experience activity: Beyond marketing 
Customer experience management is not confined to the marketing department. Full-scale journey management must include all customer-facing and customer-affecting activity, including fraud detection, pricing, credit and collections. All these functions require unified customer profiles and benefit from sophisticated analytics, real-time decisioning and access to digital events as they happen. Extending the CDP capabilities beyond marketing with the easy integration of real-time decision engines is critical for the customer experience, now and in the future.