Too often, business executives are uncertain about the quality of their organization's data. It's an important issue today in part due to regulatory issues, as well as problems with the accuracy of data drawn from disparate sources, including spreadsheets. Errors and inconsistencies lead to mistakes and lost opportunities—failed deliveries, invoicing blunders, problems with global data s...
Too often, business executives are uncertain about the quality of their organization’s data. It’s an important issue today in part due to regulatory issues, as well as problems with the accuracy of data drawn from disparate sources, including spreadsheets.
Errors and inconsistencies lead to mistakes and lost opportunities—failed deliveries, invoicing blunders, problems with global data synchronization—with estimates of the cost of errors due to unreliable and incorrect data in retail business alone as high as $40 billion annually.
Master data management (MDM) is a part of many manufacturers’ business and information management strategies. Ventana benchmarking research indicates that managing master data about customers, products, materials, vendors, charts of account, and location is critical for business success. An approach to MDM that combines business and technology components specific to the kinds of master data typically found in a given industry can help companies improve MDM deployments.
Master data includes the business objects, definitions, classifications, and terminology that, in sum, constitute business information; as well as format specifications for transactional data. MDM makes it possible to define and link master data, including those definitions, references, rules, and metadata. It seeks to establish and maintain a high level of data consistency and reliability.
By these means, a company can deploy and manage processes such that each line of business, regardless of its technological expertise, is responsible for its own data and for enforcing standard practices for conducting business and analyzing information.
Why the IT shortfall?
It’s wise to select master data management (MDM) team members able and willing to contribute information, skills, and experiences that provide an appropriate mix for achieving the team’s purpose.
Manufacturers today rely on complex enterprise applications and information systems to support business processes. Typically, each application or system—whether for customer, materials, product, or services management—1) has specific functionality; 2) handles the business context of data and rules in its own fashion; and 3) stores within it the descriptors of the data. In other words, different applications handle data differently. This heterogeneity means information is inconsistent in different parts of the organization, and leads to complications in exchanging and synchronizing information.
Master data management defines master data and synchronizes definitions and rules. It is supposed to address the problem by providing needed context. However, inconsistency in the master data itself can damage its effectiveness. For example, managers have found some responses to a data request may disagree with others.
That’s why companies are seeking true and complete commonality in data. In the past two years, Ventana investigated various MDM topics in five benchmark studies. The results of these benchmarks confirm these trends, and identify MDM as a growing priority for manufacturers.
Master data types
Master data spans business areas and subject types, and each type of master data has unique elements that must be handled accordingly.
Master data spans business areas and subject types. Each type of master data has unique elements. Taken together, they encompass most business processes and management views. The following four typically are the most important.
Customer master data is a frequent starting point for organizations adopting MDM. Among elements typically needed in market-to, sold-to, ship-to, and bill-to accounts are addresses, contact names, and hierarchies. Producing consistent master data requires identification and assessment of the existing business processes in which these elements figure. Yet companies typically store customer master data in an array of systems, among them sales force automation (SFA); customer relationship management (CRM); and enterprise resources planning (ERP).
Product master data presents organizations with many of the same challenges. Our research reveals that product data is widely dispersed across organizations, and that managing it is a cross-functional responsibility. The complexities of product data are further compounded where they intersect with customer data. Eighty percent of organization executives participating in benchmark research say they are not confident of their product data’s quality. Materials (item) master data serves as a central repository for the characteristics of a company’s inventory and related items. It contains data elements such as part numbers, descriptions, specifications, and stocking codes.
Engineers and designers create parts and assign part numbers, Procurement sources new parts; suppliers may load their parts masters into the organization’s systems; and planners create their own items for bills of materials. All of this creates heterogeneous stores of materials data that are not easy to integrate. Materials master data is core for various types of supply chain management systems, including those for material requirements planning, supply chain execution, and warehouse management.
Many of today’s global enterprises were built over decades through both organic growth and mergers & acquisitions. These large organizations have amassed different versions of ERP and other legacy systems. In such cases, the materials master data and attribute information exist in multiple data forms, languages, and degrees of detail.
Vendor master data captures data about the performance of suppliers. Its basics include information about remittances, financial stability, locations, and ownership structures. In addition, attributes to measure supplier performance—such as on-time delivery and quantity shortages—are increasingly common. Proof of adherence to fair labor laws, insurance certifications, and compliance with environmental regulations today must be captured and managed centrally.
The MDM team
Successful MDM includes good processes—including selecting and implementing technology—for managing the information. To establish and maintain a good MDM process takes a partnership of business people, to manage the master data, and IT staff, to support business efforts across the organization. Building a successful team requires a clearly expressed mission statement, well-defined roles and responsibilities for the participants, and specific metrics to manage team performance.
To be successful, an MDM team must accomplish three tasks: 1) define the business; 2) find and select the best technologies; and 3) perform the ongoing “people” tasks related to master data management.
In practice, team management often is a shared function, and members should be given opportunities to exercise leadership when their experience and skills are appropriate. It’s wise to select team members able and willing to contribute information, skills, and experiences that provide an appropriate mix for achieving the team’s purpose. Members also need to know how to examine team and individual errors and weaknesses without making personal attacks, so the group can learn from its experiences.
You may need separate teams for different master data objects—one for suppliers, for example, another for customers, and a third for products—because each object is used in different business processes. As well, each master data object is stored in appropriate business applications, and each application requires its own domain expertise.
Ventana recommends that before beginning an MDM initiative, you choose a good place to start. That is, take account of the nature of your business and select a type of data important to it. Start small, achieving short-term wins that will gain acceptance of and support for MDM, and then spread the initiative across the enterprise.
A master data management (MDM) team must identify metrics that can be used to monitor benefits, including key performance indicators that are important to the success of a particular business. A baseline should next be established to measure performance against it, adjusting targets and setting new ones as performance progresses.
Start with a subset of items or products—for example, a single parts family or commodity group. Once proven, expand the number of item categories. Brand owners and distributors—i.e., companies that primarily sell goods produced by others—may want to tackle customer master data first since it often has structure and is well understood. On the other hand, discrete manufacturers probably should look first at materials master data, a key part of manufacturing processes.
Next, build a business case that includes cost of errors. Our benchmarking research on MDM shows most organizations don’t evaluate the financial consequences of operational errors, such as incorrect deliveries or invoicing mistakes. Such errors often can be traced directly to problems with product or item master data.
Finally, determine what metrics can help you realize the business benefits of master data management; be sure to select key performance indicators (KPI) that are important to the success of your particular business. Next, establish a baseline, measure performance against it, and then adjust targets.
Creating accurate master data and managing it consistently are challenges that no enterprise can afford to ignore. When a business runs on information, it needs a solid foundation of data. Without that, the organization’s performance and success are at risk.
