Register for "Diverse, Integrated, and Real Time: Delivering the Right Data for AI/ML Success" - Tuesday, June 18,1:00 pm EST

The Customer 360 Data Program and Cloud Connectors: Guiding Principles for Success

ABSTRACT: This blog, the third and final in a series, recommends five guiding principles for success.

Let’s say you decided to re-architect your customer 360 data program. You opted to unbundle the tasks of the monolithic customer data platform in order to engage more customers with more targeted offers. Your marketing and data teams use the following architectural elements.

  • The data warehouse, such as Snowflake or Google BigQuery, helps integrate customer data.
  • The customer data platform, such as Lytics Decision Engine or Segment, helps build customer profiles.
  • Cloud connectors such as Lytics Cloud Connect help activate profiles for targets—for example, digital advertising platforms such as Google Ads and Instagram.
  • Streaming pipelines, using tools from vendors such as Tibco and Fivetran, help synchronize data across touchpoints.
  • BI tools such as Qlik and Tableau, as well as ML models, help analyze customer actions and program results.

To achieve the right results, you need to adopt the right guiding principles. This blog, the third and final in a series, recommends guiding principles for success. The first blog defined this new architectural approach, with digital advertising as a case study, and the second blog offered criteria to evaluate the addition of cloud connectors to their architecture.

The five guiding principles are to (1) start with business goals, (2) define use cases, (3) maintain a single version of customer truth, (4) take incremental steps, and (5) plan for growth. Let’s explore each in turn.

1. Start with business objectives and requirements. This principle, while self-evident, deserves emphasis because many enterprises side-step it and leap to execution mode—which undermines the C360 program’s value. To build a successful C360 program, marketing leaders and other executives first need to define its strategic business objective. The objective might be to increase revenue per customer within a target demographic, take market share from a top competitor, or reduce overall customer churn.

The strategic business objective shapes the program requirements. Suppose a cellular phone carrier sets the objective of taking market share from a top competitor. With this objective as a guide, its marketing and sales leaders define three supporting business requirements: increasing competitive sales wins, increasing renewal rates, and reducing competitive sales losses. With these business requirements as a guide, marketing and data leaders can define their use cases.

2. Define the use cases that will shape the 360 data architecture. Use cases span customer campaigns—for example, marketing, advertising, sales, or customer service—as well as analytics initiatives that report on campaign performance, segment customers, define product requirements, or personalize content. Marketing and data teams should define use cases that achieve strategic business objectives and meet business requirements. These use cases then can guide architectural decisions.

Our cellular phone carrier, for example, might scope three use cases to take market share from their top competitor.

  • To increase competitive sales wins, they publish digital advertisements that offer incentives to the other carrier’s customers.
  • To increase renewal rates, they email early and frequent upgrade offers to existing customers.
  • To reduce competitive sales losses, they publish digital advertisements that compare their satisfaction and uptime ratings to the other carrier. They also email upgrade offers to existing customers.

With these use cases as a guide, marketing and data teams can assemble the right data sources, targets, and profiles.

3. Maintain a single version of customer truth. Unless implemented well, this new architectural approach risks creating contradictory and duplicative data copies that undermine program performance and governance. To minimize this risk, data engineers must continuously synchronize updates. This helps maintain live, current profiles—the single version of truth—in the data warehouse. They must configure, manage, and monitor streaming data pipelines or streaming features within, their various architectural elements to keep the profiles up to date. 

For example, when data engineers integrate new data in the data warehouse, they must synchronize it with the profiles in the customer data platform. Then the customer data platform must store the revised profiles back in the data warehouse. When customers respond to digital ads or other target platforms, the cloud connectors must synchronize those updates with profiles in a similar fashion. Streaming data pipelines enable this synchronization to take place.

Customer 360 Data Synchronization

4. Take incremental steps. Many C360 programs fail because they try to achieve too much, too fast, then fall short of unreasonable expectations and lose momentum. Marketing and data teams can avoid this trap by plotting a series of incremental steps, each with reasonable near-term milestones for business results. If they can notch a modest but quick win early on, they gain confidence, executive support, and budget to tackle larger projects.

Our cellular carrier, for example, might first plan to take market share from their top competitor in the southwest region of the United States. Their marketing and data teams test multiple versions of digital ads and customer emails. They compare version results according to their impact on competitive sales win rates, renewal rates, and competitive sales losses. They apply the most effective versions to phase 2 in the rest of the United States, then assess and adapt further to execute phase 3 in Europe.

5. Plan for growth. Both the supply of customer data and demand for it continue to rise across industries. Enterprises that cannot support this growth reduce the value of their C360 program and even risk overall program failure. Data teams must design an open and elastic architecture that can scale with the needs of the business. They should maintain open application programming interfaces (APIs) and adaptable data pipelines that accommodate new sources, targets, and formats. They also should adopt elastic cloud infrastructure that can scale as needed, while still setting alerts and thresholds to control cost.

These principles are neither new nor unique to C360 programs. But they can provide enterprises with the right guardrails to ensure people, process, and technology come together to drive business results.

Kevin Petrie

Kevin is the VP of Research at BARC US, where he writes and speaks about the intersection of AI, analytics, and data management. For nearly three decades Kevin has deciphered...

More About Kevin Petrie