Chapter 16 Data Management Organization and Role Expectations

1. Understand existing organizational and cultural norms

The data management organization should be consistent with the company's organizational hierarchy and resources. Finding the right people requires understanding the functional and political role of data management within the organization. The goal should be to engage different business stakeholders across functions. Need to do:

Identify employees currently performing data management functions, get to know them and invite them to participate first. Only consider investing more resources as your data management and governance needs grow.
Examine the organization's approach to managing data and determine how to improve processes. Improving data management practices may require multiple changes.
Think about the organization's perspective and plan for the various changes that need to be made to better meet needs.

2. Structure of data management organization

A critical step in the design of a data management organization is determining the organization's optimal operating model. A reliable operating model helps an organization establish accountability, ensures the right functions within the organization are represented, facilitates communication, and provides a process for resolving issues. An operating model forms the basis of an organization's structure, but rather than an organizational chart, where people's names are simply placed in boxes, it describes the relationships between the various components of the organization.

2.1 Decentralized operating model

In a decentralized operating model, data management functions are distributed among separate business and IT departments. Committees are the basis for mutual collaboration and do not belong to any single department. Many data management initiatives start from the ground up with the intent of unifying data management practices across the organization and thus have a decentralized structure.

Advantages: The organizational structure is relatively flat, and the data management organization is consistent with the business line or IT department.
Disadvantages: Involving too many people in governance and decision-making, implementing collaborative decision-making is often more difficult than centralizing orders.

2.2 Network operation model

Through the RACI (who is responsible, Responsible; who approves, Accountable; who is consulted, Consulted; who is informed, Informed) responsibility matrix, using a series of documented connections and responsibility systems to make decentralized informal organizations more formal and become Network mode.

Advantages: Similar to the decentralized model (flat structure, consistent concepts, rapid establishment). Adopting RACI helps establish accountability without affecting the organizational structure.
Disadvantages: RACI-related expectations need to be maintained and enforced.

2.3 Centralized operation model

The most formal and mature data management operating model is the centralized operating model. All work is controlled by the data center organization. Those involved in data governance and data management report directly to the data management lead who is responsible for governance, management responsibilities, metadata management, data quality management, master and reference data management, data architecture, business analysis, etc.

Pros: Establishes a formal management position for data management or data governance, with a final decision-maker. Decision-making is easier because responsibilities are clear. Within an organization, data can be managed separately by different business types or business topics.
Disadvantages: Implementing a centralized model often requires significant organizational changes. Formally separating the role of data management from core business processes risks a gradual loss of business knowledge.

2.4 Hybrid operating model

The hybrid operating model contains characteristics of both decentralized and centralized models. In a hybrid model, a centralized data management center of excellence works with decentralized business unit teams, often through an executive steering committee representing key business units and a series of technical working groups focused on specific issues.

The beauty of a hybrid model is that it allows for appropriate direction from the top of the organization, and there is an executive responsible for data management or data governance. Business teams have broad responsibilities and can be given more attention through business prioritization. They benefit from the support of this dedicated Data Management Center of Excellence, helping to focus on specific challenges.

Challenges with this model include organizational set-up, which often requires additional staffing of centers of excellence. Business teams may have different priorities that need to be managed from the perspective of the business itself.

2.5 Federal operating model

As a variation of the hybrid operating model, the federated model provides an additional layer of centralization/decentralization that is often required in large global enterprises.

The federated model provides a centralized strategy with decentralized execution. An executive leader responsible for data management across the organization, responsible for managing the Enterprise Center of Excellence. This model enables organizations to prioritize based on specific data entities, departmental challenges, or regional priorities.

The main disadvantage of this model is that it is more complex to manage. There are too many layers to it, requiring a balance between line-of-business autonomy and the needs of the enterprise, which impacts enterprise priorities.

2.6 Determine the best model for your organization

The operating model is the starting point for improving data management and data governance practices. Before introducing an operating model, you need to understand how it affects the current organization and how it is likely to evolve. Because the operating model will help with the definition, approval, and enforcement of policies and procedures, it is critical to determine which operating model is best for the organization.

2.7 DMO alternatives and design considerations

Most organizations were in a decentralized model before moving to a formal data management organization (DMO). As an organization sees the impact of data quality improvements, it may have begun developing accountability through the Data Management RACI Matrix and evolved into a network model. Over time, synergies between distributed roles will become more apparent and economies of scale will be established, pulling some roles and people into organized groups, eventually morphing into a hybrid or federated model. When building an operating model, keep the following points in mind:

Determine by evaluating the current status
Link operating model to organizational structure
Consider: organizational complexity + maturity; domain complexity + maturity; scalability
Gain support from the top – a must for a sustainable development model
Ensure that any governing body is a decision-making body
Consider pilot planning and batch implementation
Focus on high-value, high-impact data domains
Use existing resources
Never take a one-size-fits-all approach

3. Critical success factors

3.1 Support from senior management

Having the right executive support ensures that stakeholders affected by data management planning receive the necessary guidance. In the process of organizational change, the new data-centric organization is effectively integrated to achieve long-term sustainable development.

3.2 Clear vision

A clear vision and a plan to drive it are critical to the success of a data management organization. Organizational leaders must ensure that all stakeholders affected by data management understand and understand what data management is, why it is important, how their work will impact management and the impact of data management on themselves.

3.3 Active change management

Managing the change process associated with establishing a data management organization requires planning, managing and sustaining change. Applying organizational change management to the establishment of a data management organization can solve the challenges people face and make it possible for the data management organization to achieve long-term sustainable development.

3.4 Consensus among leaders

Consensus among leaders ensures consistency and unified support for the data management plan and agreement on how to define success. Consensus among leaders includes leaders’ consensus on goals and data management outcomes and values, as well as consensus on the leader’s purpose.

3.5 Communicate continuously

Communication should begin as early as possible, be open and frequent. Organizations must ensure stakeholders have clear guidance on what data management is, why it is important to the company, what is changing and how behaviors need to change. You can't improve how you manage your data if you don't know what to do about it.

3.6 Stakeholder participation

Stakeholder analysis helps organizations better understand those affected by data management changes. By taking this information and mapping stakeholders based on their level of influence and interest in data management implementation within the organization, organizations can determine the best way to engage different stakeholders on the change train.

3.7 Guidance and training

Different groups require different types and levels of training, and leaders need to be clear about the direction of data management and its value to the company. Data stewards, owners, and administrators all need a deep understanding of the data stewardship program, and targeted training can enable them to function effectively.

3.8 Adopt a measurement strategy

Adopting a measurement strategy can help you understand whether your current data management roadmap is and will continue to be effective. How to develop metrics, as follows:

Whether to adopt
The degree of improvement, the increment relative to the previous state
Advantages of data management. How data management impacts solutions with measurable results.
Improved processes and projects
Identify and avoid risks
Innovative aspects of data management. How data management is fundamentally changing the way business is done
Credibility analysis

3.9 Adhere to guiding principles

Guiding principles articulate the shared values of the organization, are the basis for strategic vision and mission, and are the basis for integrated decision-making. Guiding principles form the rules, constraints, standards, and codes of conduct that an organization follows in its long-term, day-to-day life. Whether it is a decentralized operating model, a centralized operating model, or anything in between, guiding principles must be established and agreed upon to align all participants in a consistent way of doing things.

3.10 Evolution not revolution

The philosophy of "evolution, not revolution" helps minimize major changes or large-scale high-risk projects. It is important to build an organization that continues to grow and mature. Implementation of new policies and processes will be ensured and continuously improved in a manner that progressively improves data management and prioritization of business objectives. Incremental changes are easier to demonstrate and therefore easier to gain stakeholder buy-in and support, and involve those key players.

4. Establish a data management organization

4.1 Identify current data management actors

Review existing data management activities, such as who creates and manages data, and who evaluates data governance. Survey the organization to find out who may already be fulfilling the required roles and responsibilities. These individuals may hold different positions. They may be part of a dispersed organization and have not yet been identified by the business. Once you’ve compiled your “data people” inventory, identify gaps and identify what other roles and skills are needed to execute your data strategy.

Once you've completed your people inventory, assign them to appropriate roles and review their compensation to align it with data management expectations.

4.2 Identifying committee participants

Having the right people on the steering committee, keeping them informed and focused on improving data management, will help them achieve their business goals and strategic goals.

4.3 Identify and analyze stakeholders

A stakeholder is any individual or group that can influence or be influenced by the data management plan. Stakeholders can be internal or external to the organization, and stakeholder analysis can help organizations identify some of the best ways to engage actors in the data management process and give them a role in the operating model. Stakeholder analysis requires answers to the following questions:

Who will be affected by data management
How roles and responsibilities are changing
How those affected cope with the changes
What issues do people have concerns about?

The results of the analysis will identify: a list of stakeholders, their goals and priorities, and why these are important to them.

4.4 Involve stakeholders

The person or team driving the data management effort should articulate why each stakeholder is integral to the project's success. and relate the output of the data management process to their goals so they see a direct connection.

5. Communication between the data management organization and other data-related organizations

The data management organization needs to work with other groups that have a significant impact on how data is managed. These groups are usually:

chief data organization
data governance organization
Data Quality Team
Enterprise Architecture Team

5.1 Chief Data Officer (CDO)

CDO is often a member of business strategist, consultant, data quality management specialist and all-round data management ambassador. Common tasks of CDO include:

Establish an organizational data strategy
Align data-centric needs with available IT and business resources
Establish data governance standards, policies and procedures
Advise the business to enable data activism such as business analytics, big data, data quality and data technology
The importance of communicating good information management principles to internal and external stakeholders
Oversee data usage for business analytics and business intelligence.

5.2 Data governance

Data governance is the organizational framework for establishing strategies, goals, and tactics for effectively managing enterprise data. It consists of the processes, policies, organizations, and technologies required to manage and ensure the availability, availability, integrity, consistency, auditability, and security of data. The data governance process has a synergistic relationship with data management because it requires the interplay of data strategy, standards, policies, and communications.

Data governance is about "doing the right things", and data management is about "doing the data right". They are the two aspects needed to create valuable data. In this way, data governance provides direction for data management.

5.3 Data Quality

Data quality management is a critical capability for data management practices and organizations. When the goal of data quality management is to improve the quality of data shared across lines of business or applications, it typically focuses on master data management. It is common for data management organizations to grow organically through data quality initiatives because investments in improving data quality add value to the entire company, and efforts related to improving data quality can be expanded into other areas such as master data management, reference data management and metadata management.

5.4 Enterprise Architecture

The enterprise architecture team is responsible for designing and documenting the organization's overall blueprint, articulating how to achieve its strategic goals and optimize them. Enterprise architecture practices include: technical architecture, application architecture, information architecture, and business architecture.

5.5 Managing a global organization

Global organizations need to pay special attention to:

comply with standards
Synchronization process
Clear responsibility system
training and exchange
Monitor and measure effectively
Develop economies of scale
Reduce repetitive work

6. Data management role

6.1 Organizational roles

IT data management organizations provide a range of services from data, application and technology architecture to database management. A centralized data management service organization focuses on data management. The organizational team may include a data management executive, other data management managers, data architects, data analysts, data quality analysts, database administrators, data security administrators, Metadata Specialists, Data Modelers, Data Stewards, Data Warehouse Architects, Data Integration Architects, and Business Intelligence Analysts.

6.2 Personal roles

6.2.1 Executive role

Data management executives may focus on the business or technology level, while chief information officers and chief technology officers play important roles on the IT side.

6.2.2 Business roles

Business roles focus primarily on data governance functions, especially management responsibilities. The initial focus of data management responsibilities, typically defining business terms and valid values for their subject areas. In many organizations, data management specialists are also responsible for defining data attributes, maintaining data quality requirements and business rules, and helping to identify and resolve data issues, providing input to data standards, policies, and procedures.

6.2.3 IT role

IT roles include different types of architects, developers at various levels, database administrators, and a range of support roles.

data architect. Senior Analyst responsible for data architecture and data integration. Can work at an enterprise level or at a functional level. Data architects generally work on data warehouses, data marts, and their related integration processes.
Data modeler. Responsible for capturing and modeling data requirements, data definitions, business rules, data quality requirements, logical and physical data models.
Data Modeling Administrator. Responsible for the design, implementation and support of structured data assets and technical approaches to improve data access performance.
Database administrator. Responsible for the design, implementation and support of structured data assets and technical approaches to improve data access performance.
Data Security Administrator. Responsible for ensuring controlled access to data at different protection levels
Data Integration Architect. Senior data integration developer responsible for designing data integration and improving the quality of enterprise data assets
Data Integration Specialist. A software designer or developer responsible for integrating (copying, extracting, transforming, loading) data assets in a batch or near real-time manner.
Analysis/Reporting Developer. Software developers responsible for creating reporting and analytical application solutions.
application architect. Senior developer responsible for integrating application systems.
technical architect. Senior technical engineer responsible for coordinating and integrating IT infrastructure design and IT technology framework.
Technical Engineer. A senior technology analyst responsible for researching, implementing, managing, and supporting a piece of information technology infrastructure.
Desktop administrator. Responsible for handling, tracking and resolving issues related to the use of information, information systems or IT infrastructure.
IT auditor. Internal or external IT auditors whose responsibilities include auditing data quality and data security.

6.2.4 Mixed roles

Hybrid roles require both business and technical knowledge, and whether the person holding these roles reports to the IT department or the business department depends on the organization.

Data Quality Analyst. Responsible for determining data suitability and monitoring the ongoing condition of data; conducts root cause analysis of data issues and helps the organization identify business process and technology improvements that improve data quality.
Metadata Expert. Responsible for the integration, control and delivery of metadata, including management of the metadata repository.
BI architect. Senior Business Intelligence Analyst responsible for AM Intelligent User Environment design.
BI Analyst/Administrator. Responsible for supporting business personnel in the effective use of business intelligence data.
BI project manager. Responsible for coordinating BI needs and initiatives across the company and integrating them into an overall priority plan and roadmap.

DAMA Data Management Knowledge System Guide-Reading Notes 16