What CDP Use Cases Should You Focus On?

Are you considering using a Customer Data Platform (CDP) to improve your marketing in the coming year? If so, you’re probably wondering which CDP use cases should receive top priority in your business goals as you roll out your new CDP capabilities. And should you initially confine your CDP use cases to the marketing department, or quickly roll out your CDP for use in service, contact center, sales, and even logistics data support or product design? It will come as no surprise to data-driven marketers that there’s actually good data on these questions, courtesy of the CDP Institute (CDPI), which offers a free online Use Case Generator to help organizations define their CDP requirements. (You can access it here.) What can you learn from these CDP users, and their six dozen CDP use cases?

The Data Set Behind Our Report on Customer Data Platform Use Cases

How did we compile the data for our analysis of CDP use cases? Although information collected about CDP use cases in the CDPI system remains private, we recently analyzed the inputs in aggregate. The resulting report provides unique insight into how companies plan to use their CDP and what requirements these new CDP user companies expect it to meet. Here are some key findings from our research.

Finding #1: Most CDP Use Cases Are Remarkably Simple  

The Use Case Generator asks about the data types, source systems, target systems, CDP features, departments, and Key Performance Indicators (KPIs) involved in each use case.  There are about a dozen choices for each category.  Yet the average use case requires just a few from each group: about six data types, five source systems, four target systems, four departments, six KPIs, and half the available features.  About one-quarter of use cases require no more than three data sources. 

The implications are significant.  CDPs are often expected to assemble all possible data into a comprehensive unified customer view profile. This makes CDP deployment a daunting prospect and can result in a large, complex project plan – or in rejecting the project altogether because of the time, cost, and effort involved. The data tells us that less ambitious projects are feasible because CDPs can deliver value with just a few data sources, target systems, and users. 

This confirms the conventional wisdom that recommends you start with a few marketing objectives, and then gradually do more as your teams build experience and demonstrate results, an approach sometimes called “crawl-walk-run incremental deployment.”  

Finding #2: Expand CDP Use One Department at a Time

But that’s not all. The report also highlights the cost of expanding CDP use beyond a single department – since each new department will require training new users and, in most cases, connecting new data sources, data types, and target systems.  This adds concrete guidance for incremental deployment; in other words, companies should plan to deploy as many use cases as possible within a single department before moving to applications in other departments.  And when they do expand to a new department, they should again deploy many use cases in that department before moving to yet another. 

This is an especially important insight for CDPs that are intended from the start to be used across the enterprise.  For those projects, there may be substantial pressure to deploy the CDP as widely as possible as soon as possible.  Companies should recognize that early, broad deployment is inherently more difficult than narrow, deep deployment.  If quick deployment across multiple departments is unavoidable, they should be sure to budget adequate resources to meet this more demanding approach.   

Finding #3: CDP Use Cases Span All Stages of CDP Maturity  

The Use Case Generator classifies each use case by its final product. These products are arranged as a sequence that starts with unified customer profiles, next moves to analytics and predictive models, and ends with outbound campaigns, real time interactions, and cross-channel orchestration. The sequence can be considered a maturity model, since early products often provide inputs needed for later products. 

About one-third of the use cases built in the Generator were aimed at the first-stage product, assembling unified customer profiles, which can be used to create unified customer views.  Interestingly, there were relatively few that targeted the next stages of analytics or predictive models.  More aimed at outbound campaigns and real-time interactions, while relatively few were intended to reach the most advanced aim: cross-channel orchestration. 

The reason for this distribution is fairly clear, at least in hindsight: Companies are most interested in use cases that deliver measurable revenue gains, something that analytics and predictive models can only create if their results are used in campaigns or interactions. Indeed, about three-quarters of the campaign and interaction use cases included predictive models in their requirements. 

It also turns out that most analytics and predictive model use cases were expected to feed their results to something other than the marketing campaigns and interactions listed as an option in the Use Case Generator. So even those projects really had a broader goal beyond analytics or model building itself.

Once more, the lesson here is that CDPs can create value quickly: remember, one-third of the use cases set data assembly as their goal. But this value will only be realized if the project plan includes all use cases, simple ones as well as those that support other departments. Business users focused on campaigns, interactions, or other high-maturity applications may not think to include the simpler ones.  This means that project teams who want their CDP to deliver value quickly need to check that they’re present.

Finding #4: Early and Continuous User Involvement Is Key to CDP Success  

Incremental deployment focused on a single department and inclusion of profile-building use cases both seem to suggest that for customer data platform use cases, it’s only necessary to  engage a few business users at the start of a CDP project. 

But that’s just plain wrong.

A narrow set of simple customer data platform use cases may be the fastest way to gain value from the CDP, but a successful CDP will ultimately be used for many more purposes across multiple departments.  Finding a CDP that will support this later expansion depends on understanding the requirements of those future use cases when the CDP is being selected. This, in turn, depends on engaging the business users who can define those use cases from the start. 

It may seem counterproductive to involve users who will later be told they’ll have to wait to use the system. Including those users will certainly create more work at the start.  But results from the CDP Use Case Generator are clear: while most individual use cases have relatively few requirements, those requirements differ from one case to the next. In fact, while few requirements were needed by more than 80 percent of the relevant use cases, none was needed by fewer than 39 percent. 

This means the only way to build a comprehensive set of requirements is to explore the full range of future use cases at the start. Incremental deployment is not an excuse for incremental requirements definition. In fact, incremental deployment can only go smoothly if a complete vision is developed in advance.

You can download our full analysis of the Use Case Generator here.  And of course, you can also create your own use cases in the Generator itself.

David Raab
David Raab
David Raab is founder and CEO of the Customer Data Platform Institute , a vendor-neutral organization that educates marketers and technologists about customer data management. Mr. Raab has a long career as a marketing technology consultant and analyst. He coined the term Customer Data Platform in 2013. To learn more, visit the CDP Institute website.