Frequently Asked Questions About Customer Data Platforms

A CDP needs to collect customer data from disparate sources, unify them into profiles, generate insights to find opportunities and guide personalization, and send segments to activation tools. A good CDP should also:

• Give users confidence in the data with a strong identity resolution solution and transparency into profile stitching

• Democratize access across the organization with views scoped to roles and interfaces for technical and non-technical users

• Surface applicable insights about customers and put prioritization and opportunity sizes at users’ fingertips

CDPs can have major differences in their capabilities, especially if they started out with a different focus, such as providing an email service or tag management. It is not uncommon to have two or more pieces of software that identify as a CDP but perform different functions. Some of these functions include:

• Identity & foundation - focus on unifying data and creating accurate profiles that can power downstream systems

• Data pipes - emphasis on moving data around between systems quickly

• Orchestration and Customer Experience - indexed towards helping Marketers building customer journeys

Customer relationship management (CRM) tools depend on known customer data to keep track of customer data and interactions. CRMs do not aggregate data sources, deduplicate customer records, or manage anonymous data. Additionally, a CRM is typically not as well connected to the various marketing activation channels as a CDP.

Data management platforms (DMP) aggregate and categorize digital events to generate insights for ad placement; they mainly rely on third-party data. As the third-party data ecosystem erodes resulting from regulatory and privacy changes, CDPs becomes a better place to aggregate first-party data, append third-party data where appropriate, and activate the unified data directly with external systems like advertising platforms.

Master data management (MDM) tools collect and manage massive stores of data - everything from product catalogs to supplier information. Even when an MDM is solely focused on customer data use cases, they are not typically built for easy activation. Common challenges include managing the scale of today’s customer data, ensuring it is easily activated with connectors to downstream systems, and democratizing access of the data to non-technical users. 

ETL-less data ingestion and schema mapping makes integration go much faster and future additions and changes to data sources & destinations much easier. Governance and compliance to privacy laws are simplified with a single access point to manage.

Marketers are able to self-serve access to data to explore audiences and build segments. They enjoy better connectivity to activation and orchestration tools - including advertising publishers without the lag or cost of a data onboarder. 

Analysts love the accurate, trustworthy data to mine. While executive leadership loves the newfound confidence in customer metrics as a measure of business health.

In many cases customer records and profiles are inaccurate because the data is not unified. Data associated with each customer is scattered across different systems that weren’t designed to talk to one another: online, in-store, email, social media, customer service records, and more. Without unifying all the pieces and connecting the dots, it’s impossible to have a complete picture of the customer’s relationship with the brand, and any insights, predictions, or personalization will be off-base.

The data comes from different systems, so it’s often formatted differently. This creates two obstacles: one, the different bits of data have no common linking key, making it tough to connect the pieces that belong to the same individual; and two, many tools require you to manually format the data or map them to a particular schema so that they can be linked, which is time-consuming work.

Despite common language around customer data platforms and unified customer profiles, a lot of marketing tools were not purpose-built to solve a data challenge. Therefore, they are missing the key elements of a true unified customer profile:

• Not incorporating all data sources, including transactional and historical data

• Stitching of customer records is based on legacy, rule-based PII match 

• Lacking persistent identity to update customer profiles with new information and lacking transparency to build data confidence

Marketing and business teams typically can’t access customer profiles without special tooling or assistance from technical teams, which slows down building segments and launching campaigns. Customer data must be in a system that easily connects segments and audiences to downstream activation and orchestration tools.

Third-party data is decreasing in accuracy and increasing in cost due to changes in privacy laws and platform consent such as Apple’s App Tracking Transparency, CCPA, and cookie changes in browsers. Brands focusing on their first-party data strategy are finding wins today, activating and suppressing more accurate audiences and using new found knowledge of segments to seed lookalike audiences. Additionally, a CDP typically doesn’t charge for the egress of first-party data.

As the third-party ecosystem degrades with recent privacy laws and platform changes, being able to directly activate first-party audiences with a CDP has become a crucial part of the advertiser's workflow. Using high-value customers and suppressing those unlikely convert to create lookalike audiences has helped boost returns on ad spend (ROAS). Additionally, unlike data onboarders, CDPs generally do not charge to activate audiences, instead building direct connectors.

Retention, loyalty, and experience have always been associated with CDPs. Early use cases take advantage of having a highly accurate, omni-channel calculations around value and predictive elements around next best action. A retention marketer can set up a churn prevention campaign with highly personalized offers while a loyalty marketer might find a large number of customers who look like high-value loyalty members but have not yet joined the program.

Privacy and compliance around customer data, a topic that continues to get more complicated as new laws are written and new customer marketing channels emerge. For the IT professional, having an auditable and controlled customer data foundation owned by the brand is key, especially if it comes with the ability to limit PII views to only the audiences who need it.

Lack of confidence in customer data results in Data Analysts spending more time in QA and cleansing activities than answering truly insightful questions. When a Data Analyst has transparency behind unified customer views, access to SQL query the data or pull it into a BI tool and the ability to share segments and audiences directly with Marketers, they can shift their efforts to the strategic questions that will drive growth.

Customer Care is the face of your brand and the first place customers expect full knowledge of their interactions to exist. Customer Care with only a partial view of a customer's history is a risk to a brand’s identity.