8 Effective Best Practices for Master Data Governance Success

Staying competitive in the marketplace depends on being able to leverage data to maximize business performance and help your company grow. Knowing how to manage and govern different types of data effectively and efficiently is essential to achieving long-term business success. But it begs the question: how do you it?

In this piece, we’ll help set you up for success by:

• Giving you an overview of master data governance.
• Reviewing key master data management definitions and categories.
• Discussing master data governance factors.
• Providing eight best practices for master data governance that you can use to achieve business success.

Before discussing master data governance best practices, understanding what master data governance and master data management are is a must. The definitions used here are by no means all-inclusive. Still, they provide enough context to create a common understanding and glossary of terms.

Master data governance is the application of data governance factors to a subset of data called master data. The factors of data governance are about documenting definitions, sources, processes, policies, rules, metrics, and people to improve the management of data.

Master data is the data accumulated via the core business entities you use to run your business or organization. These business entities are referred to as domains of master data. Every industry uses multiple domains of master data in the execution of their organizational activities.

Some examples of master data domains in different industries include:

• Insurance companies have Agents, who sell Policies, to Customers, to cover Assets, in Locations, and Customers, file Claims, against those Assets, which are repaired by service Partners.
• Healthcare organizations have Doctors and Nurses, who provide Services, at Hospitals or Clinics to Patients, using Equipment and Medicines, purchased from Suppliers.
• Manufacturing companies buy Materials, from Suppliers, using Purchase Orders, that are charged against Cost Centers, to build Products, on Equipment, run by Employees, at manufacturing Plants.
• Public sector organizations have Agencies, in States or Cities, that provide Services, to Citizens, through Employees or Partners.

Data governance is about creating trust. Trust that data is being handled & managed correctly and meets high quality standards. Master data governance factors are what help you create the optimal conditions for delivering quality, trusted data, management and processes. Let’s look at eight key factors of master data governance that can help you create trust within your organization.

1. Master Data Definitions

Master data definitions describe business entities and their attributes to create a common definition of each domain of master data used by your organization. For example, a customer master might contain business name, email, phone number, shipping address and billing address. A product master might contain category, SKU, size, color, and material attributes. And an equipment master might contain type, model, serial number, manufacturer, and location attributes.

2. Master Data Policies

Master data policies describe internal and external regulations that must be adhered to as part of the management and use of master data. Policies can be oriented around many aspects of master data management and use. One example of a data quality policy you might create is that master data must contain the complete set of attributes in the domain definition. While a data privacy policy might state you must obtain consent for processing before you use personal information. A risk management policy might state there must be separation of duties between the creator and approver of new cost centers.

3. Master Data Rules

Master data rules define how you execute and enforce policies. Here’s an example of how it works with the policy stating you must obtain consent for processing before you use personal information. One rule might enforce the collection of consent attributes such as billing, marketing, and third-party sharing before a customer record can be created and approved. Another rule might check that the marketing consent attribute is set to “Yes” before a marketing automation system can send a customer a promotional message. It’s not unusual to have multiple rules to address the requirements of a single policy.

4. Master Data Catalog

A master data catalog documents where master data is, in applications and analytical data stores on-premises and across multiple cloud ecosystems. It also documents the domain of master data and its attributes, as well as the hierarchical and graph relationships. Understanding what master data you have is critical to ensuring its consistency across sources and the accuracy and completeness in each source. For example, if mergers and acquisitions are a part of your company’s growth strategy, being able to quickly compare master data in the acquired company’s systems to your master data definitions can reduce integration costs, accelerate business value, and reduce financial reporting risk.

5. Master Data Lineage

Master data lineage shows how master data moves across sources and is used in analytical and operational processes. Lineage is beneficial in many business activities including privacy compliance where it supports record of processing activity (ROPA). It helps you understand what data is being used, how and by whom. Artificial intelligence (AI) within business processes, such as recommendation engines or robotic process automation (RPA), benefits from master data lineage by identifying where to place an algorithm within a business process as well as the structure of data an algorithm can expect as an input. Master data lineage is also important for activities like customer onboarding in financial services, product track and trace in pharmaceuticals, and sustainable sourcing of consumer goods.

6. Master Data Stakeholders

Master data stakeholders are the people across functional areas of the business who are key to the success of managing master data. This includes two groups of people. The IT people who are responsible for the architecture and management of databases, applications, and business processes, and the business subject matter experts who create the standard master data definitions, policies, and rules. Data stewards who are responsible for remediating data quality issues for specific master data domains include legal and security people who are responsible for data privacy and protection, as well as cross-functional leaders who comprise the governance board responsible for resolving disputes between different functions within an organization.

7. Master Data Workflow

Master data workflow defines the processes used to manage master data. Workflow spans a variety of task-based processes including creation, update, and approval of master data definitions, policies, and rules, as well as creation, update, and deletion of master data records. Well defined workflows improve productivity and collaboration between stakeholders responsible for different aspects of master data management. For example, supplier onboarding requires collaboration between finance, procurement, and legal teams to ensure compliant screening, which might include parallel workflow for address and bank data verification, credit and sanctioned party list checks, and review of certificates of conformity and insurance.

8. Master Data Metrics

Master data metrics need to be defined to help you measure and manage data, process, and people. Common data metrics include the number of duplicate records in an application, as well as the accuracy and completeness of master data records. You’ll also want service level agreement (SLA) metrics for end-to-end processes to understand things like how long it takes to approve changes to a master data definition and implement those changes in the sources of master data. Metrics help you monitor the productivity and efficiency of people performing specific tasks within those end-to-end processes, such as how long data stewards for different domains or applications take to process change requests.

Now that we’ve examined the eight key factors of master data governance, let’s look at some best practices you can implement for each of the factors discussed above.

1. Focus your governance scope

Focus on the master data entities that are critical to your business processes like order to cash, record to report, procure to pay, and hire to retire. It’s likely that different systems and regions will use a different set of attributes to describe the master data entities, so try to define the minimum set of attributes that must be consistent across systems for your business processes to execute efficiently and effectively. Trying to govern too many attributes will make it difficult to get consensus from stakeholders and delay implementation.

2. Engage business policy experts

Many policies will address national, state and industry regulations that have high liability consequences. Privacy regulations like General Data Protection Regulation (EU GDPR) and California Consumer Privacy Act (CCPA), financial reporting regulations like International Financial Reporting Standards (IFRS) and Generally Accepted Accounting Principles (US GAAP), and industry regulations like BCBS 239 in banking and Open Payments (Sunshine Act) reporting in healthcare require expert level help to navigate successfully. The people working in legal, finance, risk, and audit will have the expertise needed to ensure you’re creating policies that adequately satisfy regulatory requirements.

3. Clearly define ownership and accountability

Master data rules should be approached in two parts. The first part is to define the rule — like a customer must have a valid shipping and billing address before an order can be created. The rule’s definitions should be owned by the business’ subject matter experts. The second part is the creation of executable code to enforce the rules like performing an API call to a postal address verification service. The executable code should be owned by IT because it will be implemented in business applications and data management tools.

4. Automate discovery and cataloging of master data

It’s likely you already have a data catalog or catalogs being used in your organization. This gives you the opportunity to leverage the learnings, people, processes, and technologies from those projects. With the number of master data sources and volume of records growing exponentially, search for tools that use AI and metadata to help you automate discovery and cataloging of master data. Advanced data sharing capabilities are also a must to enable compliant self-service access with full auditability of who has accessed which data.

5. Automate master data lineage and process flow mapping

Not only are the sources of master data growing exponentially, so too are the data integration and movement jobs. Tools that use AI and metadata can help you efficiently scale by automating the data lineage mapping process, and help you identify data movement processes you didn’t know about. You should also map owners of applications and data stores as part of the lineage map, as this will facilitate greater collaboration and productivity in the stewardship of master data.

6. Clearly define roles and requirements

Start by designing roles and responsibilities around the desired outcomes, not around people. If you select the seemingly obvious candidates for key positions before the roles are defined, you’ll define the roles based on the qualifications of that person. Instead, first define what technical skills and business knowledge are required for a person to be successful in the role. Then determine whether you have the right people in-house, can train people to fill those roles, or need to source them externally.

7. Streamline and optimize workflows

Work with stakeholders to understand how they currently manage master data and how the activities of different groups are connected. Evaluate how much coordination is required and what work should be designed around a highly structured workflow. Then look for ways to automate routing, prioritization, and notification to increase productivity and efficiency. For example, routing change requests to the inbox of the data steward responsible for a specific master data domain and prioritizing change requests based on service-level agreements (SLAs) for specific business applications. Be sure to keep a full audit trail of changes for every step of the workflow to speed approvals and enable rollback.

8. Measure business value

While technical data metrics are good, what you really want to demonstrate is the business value of master data management. This requires you to design a metrics hierarchy that links the data metrics to process metrics and strategic key performance indicators (KPIs). For example, accurate inventory data improves accuracy of delivery-date quotes. Accurate shipping data increases on-time delivery rates. Accurate billing contact data decreases invoice delivery time while accurate tax data decreases invoice disputes. All of these have an impact on days sales outstanding (DSO). Proving the efficacy and value of master data management will help you and your organization in the long run.

Now that you’ve got a solid understanding of master data governance key factors and best practices, expand your knowledge with our on-demand MDM and Data Governance Summit. You’ll learn about the latest best practices directly from customers, industry analysts and Informatica product experts.

Want a refresher instead? Register for our on-demand webinar, Back to Basics: What is Master Data Governance? It explores ways to create master data governance success and MDM at your organization.