Customer data is one of, if not the most, valuable assets for a company. McKinsey & Co. finds that organizations that leverage insights from quality customer data outperform peers by 85% in sales growth and 25%+ in gross margin. Customer data informs upsell strategies, targeted marketing, excellent customer support, etc. It’s also a key component in large digital transformations. As such, lots of time, energy, and resources have been spent trying to maximize customer data’s value. One method for unlocking the value of customer data, and ensuring it’s quality and usability for everyone within the organization is customer master data management.
This article explains what customer master data management is, why it is important, and the key components for addressing it at enterprise scale.
Customer MDM seeks to create a unified, accurate, and persistent set of identifiers and attributes that describe a customer and that can be used to connect customer data across multiple organizational silos, and business processes and units. This mastered data, that is continuously ‘live’ and up-to-date, can be fed to operational and analytical systems to drive business outcomes.
Customer data mastering is achieved first by consistently and continuously integrating customer data across various enterprise sources to track the journey of the customer from lead to sale to after-sale service. This requires customer records across systems to be matched and unified in a cluster, providing a golden record customer view and unique ID for persistent customer 360.
Why Customer Data Matters
1. No single view of customers: Lack of a single source of truth results in business-wide decision making that is not driven by robust customer insights - business development teams lack relevant information to convert opportunities; sales teams do not know enough to inform expansion strategy or prioritization; marketing teams cannot recommend actions that are backed by most up-to-date customer interactions & behavior; risk and compliance teams struggle to meet know-your-customer (KYC) standards; customers cannot have a unified customer service experience due to various disparate un-connected accounts with the same vendor
2. Persistent data errors: Errors can result in unfavorable sales outcomes and unnecessary spending in fixing mistakes ranging from incorrect customer deliveries to bounce back of marketing campaigns. Sales teams spend valuable time on fixing errors and responding to customer complaints that could be spent on expansion. Inaccurate data also slows down lead conversion with increased difficulty reaching out to customers with wrong or outdated contact information.
3. No insight into corporate hierarchies: More pertinent for B2B business, the inability to place a customer within its corporate hierarchy restricts the benefits of consolidating information across business units and functions. Lack of hierarchy information limits the effectiveness of decision-making in the customer journey from lead conversion all the way to customer retention and growth efforts. In some cases, customer grouping requirements can be met by creating reports based on attributes (e.g. a filter in a data visualization tool), but often, the business requires a more consistent structure for understanding complex customer relationships to be reflected in the mastered data.
What are common data sources for Customer Master Data?
While the list of internal and external customer data sources is long and varied at the enterprise level, common key sources of customer data are:
ERPs e.g. SAP, Oracle
CRMs e.g. Salesforce, Microsoft Dynamics
Marketing automation tools e.g. Adobe Experience Cloud, Eloqua
Components of Customer Mastering
Persistent IDs: Creating a unique ID that connects customer records within and across systems and enables a golden record view of customers, grouping the underlying records to ensure accurate, consistent information feeds a holistic customer view.
Deduplication: Identifying where there are erroneous or necessary duplicate customer records to ensure that the same underlying customer is treated as one
Data aggregation: Joining customer information across disparate systems and business units to provide a truly holistic view of interactions. Customer data can be continuously integrated with operational and analytical systems so changes in customer information and behavior is seamlessly flowed through the enterprise to reflect in business reporting and actions.
An ultimate Customer Master Data set should be a high-priority goal for the organization hoping to best leverage their digital assets in line with key DataOps principles.
Technical features of modern customer mastering solutions
Machine learning-first approach: Traditional rule-based data mastering (at the most basic level, ‘if-then’ type statements) is not scalable given the volume and diversity of customer data that the modern enterprise creates and manages. Rules require consistent high manual effort from creation to maintenance and become a complicated web to untangle for any data team once scale is reached. By leveraging a machine learning approach, it allows customer data to be mastered with a fraction of the manual effort while maintaining team input and influence on how the data should look. Poor underlying data quality can mean piecing together partial information like half-completed address fields - a task built for a machine learning approach.
Hierarchy mastering and classification: Flexibility to classify customers based on how you manage customers, from sites and sales regions to corporate hierarchies. It’s critical that mastered customer accounts are able to link accounts at an overall hierarchy level.
Data enrichment: Built-in pipeline enrichment capabilities to ensure critical customer information such as address, email and URL are kept up to date. While information validation is now standard practice for many, it’s vital to establish a streamlined process as part of the data pipeline.
How is Customer Master Data Management Different for B2B and B2C Enterprises?
While master data extends to both individual (B2C) and business-focused (B2B) customers, there are nuances in how each is addressed.
Typically, customer name comparison for business customers is more complicated e.g. J&J vs. Johnson and Johnson. Often, the value of a machine learning approach to mastering becomes evident as it is much more difficult to accommodate for corporate name variation using a rules-based simple string comparison. While individuals typically change their names infrequently or not at all, it is common for organizations to change names during the lifecycle of the business or due to a trigger event, such as an acquisition. This increases the need for regular validation of base customer information.
The availability of external customer data also differs in nature for B2B and B2C enterprises. There are two key benefits incorporating external data sources into customer master data management - (1) providing a richer view of the customer and (2) validating that your current information on the customer is up to date (reference data). While the world of externally sourced data on individuals continues to expand, and regulation aims to keep pace, structured external data sources are much more common for business customers. The ease of googling the address of Nestle corporation versus the (hopeful!) difficulty of finding search results for your personal address. Relationship management also varies. B2B customers often involve more structured and complicated cross-party relationships in the form of hierarchies.
Un-mastered customer data is in most cases the primary obstacle in the way of B2B organizations making data-driven customer-facing business decisions.
Disparate data that sits across business units, and product and geographical boundaries weigh down the organization by presenting a ‘blinders-on’ view of the customer challenge. For instance, sales teams often lack visibility of the true customer hierarchy (location, corporate parent, etc.) to be able to up-sell, offer better customer service, and save money from better contract negotiation. With mastered data, B2B businesses can better manage interactions with multiple buyers within a customer in a single view, limiting unnecessary roadblocks and steps to making the final sale.
Operations can be streamlined as data across multiple systems, including legacy deployments, can be unified with the creation of a golden record across the organization. Golden records can in turn drive new insights through novel reporting and views that may not have been possible without mastering. To further enhance operations and analytics, mastered data can also be published across a range of stores with changes and updates reflected throughout.