By Ian Clayton, Chief Product Officer, Redpoint Global
The modern consumer demands a relevant, personalized customer experience (CX) irrespective of channel or how a customer’s journey unfolds. As one example, an option to buy an item online and have a store associate deliver it to your vehicle an hour later was unheard of just a few years ago, and it is now table stakes. More recently, generative AI (GenAI) went from being a novelty to a potential CX gamechanger seemingly overnight.
Enterprise companies that invest in customer data platforms (CDPs) to power personalized experiences through the creation and activation of a unified customer profile across all touchpoints are challenged to keep up with the rapid change. As defined by the CDP Institute (CDPI), a CDP is a system that collects, unifies and analyzes all your customer data from various sources, creating a Golden Record (a “Customer 360” or “Unified Customer Profile”) that is a single source of truth for all your customer information. The conventional viewpoint holds that a CDP is an all-in-one packaged solution where a single vendor provides everything that is needed to create a Golden Record, offering the CDP as a SaaS and controlling the data in its own database.
According to research from CDPI, CDP industry growth has slowed in recent periods. While several reasons are given – including an overall shrinking of tech investment to make room for new spending on AI – one reason is because an intractable, packaged solution limits the flexibility of marketers and business users to quickly pivot to meet new CX use cases, or to assemble their marketing technology stack according to their business’s unique requirements.
This became the catalyst for the composable CDP, which provides unprecedented flexibility and agility, while also solving for the problem of keeping up with rapidly evolving customer journeys and expectations for a seamless CX. In a composable CDP the customer records reside in a modern data cloud and the enterprise brings together best-of-breed capabilities from one or more vendors to create a purpose-built, cohesive platform. A composable CDP provides a company with the best of both worlds – control as well as the flexibility to add or change components as the business and use cases evolve.
These components might include data ingestion tools, data hygiene, data observability, identity resolution, segmentation and activation systems, real-time interaction modules, and more. The key is that these pieces can be swapped out or customized as requirements and business needs evolve.
What is a Composable CDP?
One reason for the surge in popularity of the composable CDP is that the technology is closely associated with the adoption of a modern data cloud as a single source of truth for customer data. A composable CDP that runs in a data cloud is referred to as a “zero-copy data” or a “data-in-place” CDP, and is built around the principle that all customer data remains in a centralized data cloud or data lake (e.g., Snowflake, Google BigQuery, Amazon Redshift, etc.). The data does not leave this central repository; instead, tools and applications access and process the data where it resides. This approach is becoming popular as companies seek to leverage existing investments in data infrastructure without duplicating or moving data.
One key distinction between a packaged CDP with a built-in database and a composable CDP in a data cloud environment is that in the latter the enterprise maintains the data and can choose the components it wants to build unified customer profiles. Added flexibility extends to being able to connect to all enterprise data sources and MarTech touchpoints.
While the notion of a composable CDP is closely intertwined with a data cloud, a composable CDP does not inherently require a data cloud. A composable CDP can be deployed on-premises, in a private cloud and it can also run in a SaaS environment and maintain composability. In this latter configuration, the software vendor will control the built-in database, management, security and operational components, but will give the marketer or business user the ability to select only the functionality that is needed, without a separate sign-on for each process.
A “build vs. buy” decision has been co-opted to mean SaaS or “data-in-place” CDP, but it can also mean SaaS vs. an on-premises or private cloud environment. That is, whereas most products to support a composable CDP are provided by vendors who specialize in a particular function – such as reverse ETL – some conventional CDP vendors have unbundled parts of their system to support a data cloud-based approach, even when the parts can be assembled on-premises and in a private cloud.
The concept of a composable CDP environment is about more than where the data resides – it is about enterprise choice over what the assembly of those unbundled parts looks like.
Keep the “CDP” in Mind in a “Composable CDP”
A composable CDP makes it easier to create streamlined connections to a company’s preferred point solutions, and change those connections at will, but it is the freedom to assemble a CDP using best-of-breed solutions should not preempt the reason for investing in a CDP in the first place, which is to ultimately gather a deeper understanding of a customer, household or other entity and use that knowledge to power an exemplary CX.
A composable CDP, in other words, goes beyond simply moving data from point A to point B – extracting data from a data warehouse or data lake and syncing it back into operational tools, such as CRMs, marketing platforms or customer support systems. Reverse ETL might be part of the ecosystem, but a complete, composable CDP also includes first-party and third-party data ingestion, cleansing, enrichment, identity resolution, and real-time data processing capabilities.
The most important way a composable CDP enhances CX is by retaining core CDP functionality, i.e., ingesting and storing enterprise data from every source and making data ready for business use the instant it enters the system. Automated data cleansing, data enrichment, matching and merging using advanced identity resolution are vital for creating a Golden Record that reflects an accurate, real time understanding of a customer. If that core functionality becomes the responsibility of a mismatch of different vendors, a CDP buyer must weigh potential cost savings against any degradation of the unified profile.
The bottom line is that for a composable CDP to retain the key elements of what it means to be a CDP, it is not enough to just assemble vendor components on top of existing data. The enterprise needs assurance that the data is made ready for business use.
Data Quality in a Composable CDP Unlocks Innovation
A composable CDP that prioritizes data quality in the building of a Golden Record ensures that all downstream processes and components are working with the same unified customer profile. Data is never siloed by process or channel, providing marketers with confidence that other CDP functions – segmentation, activation, orchestration – are based on the absolute most recent understanding of the customer.
High quality data across the complete data lifecycle combined with the flexibility to switch out components in a seamless connection to all your end channels enables marketers to quickly adapt to new and emerging use cases and technology trends. A few years ago curbside pickup became the latest CX craze. Today companies are scrambling to integrate AI and GenAI capabilities – and high-quality data is mandatory for high-quality AI results. With a composable CDP, the enterprise can adapt quickly without an expensive rip and replace. If the enterprise wants to swap out an analytics system, it is free to do so. Not bound to a vendor requirement, it is never necessary to duplicate an existing component of your tech stack.
Marketers are thus able to more easily experiment with new strategies and tactics for enhancing a personalized CX because they are not locked into a single approach.
The Role of Data Governance in a Composable CDP
The more flexible approach to data management in a composable CDP also supports better oversight of data integrity, security and compliance. Data governance is made easier due to the fact that data remains in place within a data cloud, allowing customers to maintain a single point of data governance, i.e., no disparate data siloes required to create Golden Records.
The maintaining of high-quality data across all touchpoints ensures not only the accuracy and completeness of Golden Records, but contributes to the overall integrity of enterprise data, reducing the risk of errors and inconsistencies. A composable CDP may also include data observability capabilities, or integrate with external monitoring tools, to provide real time visibility into the health of enterprise data within the composable ecosystem. This feature improves overall data governance by making it easier to enforce data management policies and respond quickly to any issues.
By leveraging the composable nature of the CDP, organizations can better align their data management policies with their business needs, ensuring a secure, compliant, and high-integrity customer data ecosystem.
Clean, comprehensive data and an adaptable, customized architecture that does not add complexity to a business’s processes or technical setup are the chief benefits of a composable CDP environment. With a composable CDP, enterprise companies retain control and flexibility in meeting the unique business, data and CX use case requirements of the business.
Ian Clayton is a technology professional with more than 20 years of expertise in the software and marketing automation industries. As Chief Product Officer, Ian is responsible for driving strategy and development of the Redpoint Global products. He previously served as VP of Engineering and Ian was a key contributor on the product, technical and engineering teams at Experian, ClarityBlue, and Honeywell prior to joining Redpoint.