Business strategy achievement requires data management capabilities. Define these first.
Data management enables the storing of everything from genomic data to Xbox scores to your Pandora playlists. If the data were unified, we’d have the beginning of master data management.
Organizations have data throughout their environment that provide single views of key data entities common across their organization. Data management provides a single view of data. Master data management provides a complete view of your organization’s data.
Broadly speaking, civilization has witnessed five generations of data management following manual processing using paper and pencil:
- Mechanical punched card: data processing
- Stored program: sequential record processing
- Online network: navigational set processing
- Nonprocedural: relational databases and client-service computing
- Multimedia databases: object-relational databases with relationships
Data models, scaling, automation, integration, and workflows increase the complexity of generating usable information from data.
Technology leaders who are thinking ahead must answer three questions to stay competitive:
- Why is master data management the backbone of an organization?
- What capabilities are required for business-strategy achievement?
- How do these capabilities translate into tangible roles within my organization?
The business case for master data management
Master data management maximizes business outcomes with improved data integrity, visibility, and accuracy. The result is better decision-making. The efficiency and effectiveness of decisions are at the heart of every organization. Are you deciding on the best location for that off-site meeting? You need data. A list of the top 1,000 venues is interesting, but a cross-section of the top ten sites—as ranked by attendees over the last three years—is more useful. Are you developing your business strategy? A summary of 100 business cases with corresponding business strategies is useful, but a revised view of only business strategies that were successful provides more meaningful information.
We collect data. We assemble information. We create knowledge. It’s knowledge that we’re striving to generate. To get there, we need people, processes, and tools to enable the best decision-making possible.
Better decision-making, reduced operational friction, and repeatable processes all benefit from understanding how your organization values and utilizes information. Achievement requires a master data-management program.
We’re talking about capabilities
Competencies and capabilities are different. Competencies measure how a company can deploy resources and use them to achieve business strategies. Capabilities, on the other hand, are the abilities, resources, activities, routines, and processes to build a competitive advantage. Competencies are skills. Capabilities are abilities.
Here’s another way to delineate between competencies and capabilities. Competencies are individual characteristics and capabilities are organizational. Let’s address the organizational elements.
An organization’s capabilities are core functions or the secret ingredients for success.
Master data management has three, high-level capabilities: business capabilities, information-services capabilities, and data-management capabilities.
- Governance: the political process of changing organizational behavior by an established system of who has the right to make decisions
- Stewardship: business ownership of data quality for one or more subject areas; deduplication; maintaining hierarchies; and developing business rules
- Platform and architecture: technology and data-management assets including data modeling, data architecture, and metadata management (data dictionaries, glossaries, and data lineage)
- Security: data availability, protection, disaster recovery, and data redundancy
- Intelligence: ad-hoc query and real-time dashboard capabilities; making the data usable
- Analytics and visualization: core reporting, advanced analytics and risk management, regulatory and statutory reporting
- Workflow: process-model data flows
- Quality: dimensions of data quality
- Integration: model connection interfaces to entities
- Operations: operational transactions and business processes of the enterprise
- Data acquisition: ELT, audit, balance and control, and testing
- Curation: the active, ongoing management of data throughout its lifecycle from creation to archiving or deletion.
- Science: data mining; establishment of methods, processes, algorithms, and systems to extract knowledge or insights from data in various forms, either structured or unstructured
- Performance: enterprise performance management of thresholds and tolerances
Design of progressive data management programs accounts for the social, business, and technological changes that can affect how data is managed throughout an organization. Stay focused on which specific organizational capabilities will be required for your master data-management program to provide better insights into your data.
The roles of data management
Despite your best efforts, eventually the conversation will shift to who’s doing what to support the necessary data activities. The roles below are included as illustrative models of potential general role descriptions that address the majority of organizational data-management activities. Roles can be compacted if teams are lean or expanded if organizational needs are large.
- Data architect: identifies objects and data elements to be managed, specifies the policies and business rules for how master data is created and maintained, describes any hierarchies, taxonomies, or other relationships important to organizing or classifying objects, and explicitly assigns data-stewardship responsibility to individuals and organizations
- Data custodian: has ownership of the data, maintains accuracy and currency of the assigned data, and determines the security classification level of the data
- Data steward: implements data policies, standards, procedures, and guidelines concerning data access and management
- Data business analyst: collects, manipulates, and analyzes data
- Project manager: appoints and supports data stewards in their areas of responsibility
- Business relationship manager: determines which data requests will be queued and executed
- Business intelligence specialist: serves as the business and technical subject-matter expert on data or information assets
- Database administrator: is responsible for storage, organization, capacity planning, installation, configuration, database design, migration, performance monitoring, security, and troubleshooting as well as backup and data recovery
- Data scientist: applies knowledge and skills to conduct sophisticated and systematic analyses of data to produce insights
- Data engineer: develops, constructs, tests, and maintains architectures such as databases and large-scale data-processing systems; integrates, consolidates, and cleanses data
- Data developer: develops, tests, improves, and maintains new and existing databases to help users retrieve data effectively
Don’t assume personnel are clear on their responsibilities. First, create each job description. Second, validate these job descriptions within the organization to ensure that gaps and overlap are addressed. Third, develop a RACI-accountable and responsible metric to assign ownership. Fourth, develop job postings. This is the job description jazzed up to represent the flavor of the organization and the team where the role resides.
Lastly, many roles require training. Data stewards and data custodians immediately come to mind. These roles have specific functions to perform. However, it’s not sufficient to only train folks in new roles. The organization as a whole must be educated to drive the change collectively. Master data-management isn’t a separate movement; the change needs to be organizational.