Maturing into a CDP (data governance that aids CDP adoption)

We’ve written before about how to future-proof your CDP initiative, and those posts have generated a lot of questions from clients about how they should get started. While much reflection by digital marketers focuses on determining which CDP is the right one for their enterprise, the quality of data that would ultimately go into that system is often overlooked. That’s unfortunate. To paraphrase Socrates, unexamined data is not worth using.

When Credera begins to take a client up the CDP Maturity Curve, we start with data governance, investigating each of what we call the Four Data Governance Pillars.

Data Ownership

It may be surprising, but when we look at a source of first-party data for clients and we ask who in the enterprise is accountable for the data, we sometimes are met with a shrug. Department A may say this is B’s responsibility, and Department B will point to A or C. That’s not a good sign. Only when there is accountability can you govern the quality of the data.

Well before you proceed along the path to CDP Maturity, confirm who has the budget and resources to make improvements.

Data Quality

The second law of thermodynamics makes no exception for data destined to feed a CDP. It states that everything moves, over time, toward entropy. Like a child’s bedroom, it doesn’t get tidier over time! This Data Governance pillar focuses on rules and validation. The rules are for how the data is captured and refined, and validation asks how data quality is affirmed periodically.

Every source of customer data should be subjected to monthly or semi-annual review and include a path toward resolution for issues found. Examples of validation tools include:

• IBM’s DataStage and QualityStage

• FirstEigen’s DataBuck

Data Management

Management of data can be thought of as the ways the data is used upstream of a CDP. Master Data Management systems (MDMs) include data lakes and warehouses, where data is unified and secured and from which insights can be drawn through dashboards and other data visualization tools.

During unification, metadata management processes ensure uniformity. For instance, if one feeding system stores U.S. records as “United States of America” and other as “USA,” the systems are unified over a common field value. Here are some other common use cases:

• Incorrect or conflicting formats of attributes, notably in date fields: Date of Birth, Date of Joining, etc.

• Similarly, conflicting logical values (1 v 0, True v False, Yes v No)

As you can guess, the integrity of metadata management contributes to the CDP-readiness of that source.

Data Security and Privacy

Customers want to know their data is used only in the ways they have approved. Equally important is the assurance that their data cannot be stolen. Both aspects are managed through systems like LDAPs (lightweight directory access protocols) such as Okta and Consent Management Platforms (CMPs) such as OneTrust.

Data privacy of the data flowing into a CDP is essential, because the CDP can often generate personalized experiences that, if strict governance is not in place, can telegraph to customers that their privacy wishes are not being respected.

The Bottom Line

If there is a single word to summarize what these four pillars represent, it’s “trust.” Can you trust that the data going into your CDP is a true picture of reality and that the insights generated by CDP reporting systems are reliable? Can your customers trust you to be good stewards of their behavioral, transactional, and customer service data?

These data governance pillars can help achieve that high level of control. Making sure each is in place before you move to a CDP future will save you time and money and maximize the return on investment for every use case your CDP addresses.