Promote data sharing and break organizational silos.
Effective data management enables data sharing and access across the enterprise which can eliminate data redundancies and lead to overall improved data quality and provide better services. Additionally, data analytics and emerging technologies are dependent on having strong and acceptable data management functions in place, so you have quality data with which to work. The CDO role is to break down organizational silos to foster better transparency and collaboration into the data needed to succeed in transformation projects.
2.1 Collaborate with your CIO Office, statistical official, and evaluation officer (if present in your agency) to understand the history of your agency’s data and infrastructure maturity.
2.2 Identify a strategy to optimally organize and assign data stewards across the enterprise.
2.3 Provide opportunities to train and empower data stewards to be effective in their role.
2.4 Establish an automated means to manage stewardship responsibilities.
2.5 Collaborate with data stewards to develop data lifecycle management best practices. The OPEN Government Data Act mandates that CDOs establish data management best practices to improve the quality of agency data.
2.6 Determine how your data management program will be organized. The data management domains of your program will serve as inputs to your data management strategy (see next step).
2.7 Develop a data management strategy in collaboration with your CIO, Chief Technology Officer (CTO), Chief Enterprise Architect or Chief Data Architect, and agency leadership, based on the findings and inputs from earlier action steps. A data management strategy will help you think through how to prioritize investments and resource allocations (e.g., data analytics, data infrastructure).
2.8 Develop an Organizational Change Management (OCM) plan to support your data management strategy. Integrating change management strategies can successfully persuade stakeholders to implement or adjust data management methods or practices.
2.9 Prioritize data sharing projects that generate measurable business outcomes and align with strategic objectives. Using an incremental approach to data management not only delivers benefits to impacted business stakeholders but it also helps you gain quick wins.
2.10 Collaborate with data stewards to develop common business language, such as data term definitions for metadata. Managing metadata vocabularies, formats, and standards ensures a common understanding across the agency of data meaning and its uses.
2.11 Identify data consolidation opportunities across database systems including data sharing and access between business lines. Evaluating data management and analytics capabilities can help you identify open source or shared service opportunities across systems to improve cost efficiency.
2.12 Design a conceptual target data architecture and enterprise solution for databases and other data platforms creating an environment to establish formal data standards. Establishing a target data architecture brings your strategy and the technology side together to show the relationships between your agency’s data functions, technologies, and types.
2.13 Begin data aggregation activities for data consolidation, integration, and service projects. This action ensures you have compiled and processed all the data from different databases needed for your data projects.
2.14 Define and track performance metrics across your data management program functions and monitor overall progress toward them. The information collected provides feedback and inputs into your data management strategy.
2.15 Routinely assess all agency data management tools and technologies (e.g., data analytics capabilities) including Operations and Maintenance (O&M) costs for efficiencies. With the evolution of data tools, it is important to constantly evaluate your existing resources to determine where cost savings opportunities are.
Build in-depth knowledge of the agency’s data history and infrastructure maturity.
Gain familiarity with the business cases for the agency’s planned investments to understand context and set expectations before engaging with individual agency offices.
Conduct knowledge transfer meetings to help uncover operational silos and determine data sharing needs of stakeholders who perform data management functions. Gather available documentation on existing data systems and their IT environments (e.g., the IT infrastructure).
Implement a data stewardship model to break silos across agency offices and create partnerships between the CDO and agency offices as well as amongst data stewards themselves.
Organize stewards across your organization to establish accountability and responsibility for the processes that ensure effective control and use of data. As defined in # Play #1, first identify all the data stewards across various business lines performing data management and analytics functions.
Engage with these stakeholders to build consensus on a standard set of core principles and clear expectations of stewards.
Facilitate cross-sharing of best practices for data usage by having meetings with informal data stewards to discuss the activities they perform in managing their data.
Work iteratively with informal data stewards to develop a stewardship framework based on their input and feedback, and gain consensus on their role (e.g., tracking data inventories and collections.
Coordinate with data stewards to identify the linkages needed between datasets, and to lead cross organizational projects. This type of framework empowers data stewards by considering their needs and opinions and fosters knowledge sharing.
Gain additional insights across silos of operations and support any future data standardization projects.
Document all existing data assets to show the value of agency data.
Engage with data stewards to obtain and document business requirements and use cases for their data systems.
Obtain relevant metadata and classify the types of data being collected, used, and stored which will require an automated means for stewards to interface and accumulate such data. Be sure to provide guidance for sharing and/or obfuscating/tokenizing sensitive data. Developing a metadata inventory is important for building trust and enabling data sharing.
Implement a data management program to develop decision making policies and procedures.
Raise agency data awareness early by developing best practices for managing agency data in partnership with data stewards.
Develop initial data standards on the representation of data to educate agency stakeholders about the value of data management, standards, architecture, and data driven decision making.
Create a data management framework connecting your agency’s data management program to the needs of your stakeholders.
Include senior leadership and the agency data analytics community as stakeholders so that data management can best inform analytics.
Implement data management frameworks which include specific disciplines, policies, and competencies that are applied throughout the data lifecycle.
Build agency capacity to establish formal data management policies and standards on implemented data assets.
Implement your data priorities with a focused data management strategy that includes iterative approaches to addressing the various challenges of enterprise data management.
Consider including collaborative activities and incremental data projects.
Align your agency’s IT strategy and modernization initiatives to the data management strategy to ensure IT and data are coordinated.
Explain how data analytics fits into your strategy.
Most importantly, engage senior and executive leadership from across business lines to garner support for your enterprise data management vision which ensures you have the backing from across the organization.
Develop an OCM coupled with project management and system engineering approaches to drive successful, largescale transformations (e.g., IT or data modernization, workforce strategy, and policy changes).
Integrate OCM strategies into data projects to advance strategic relations, increase teamwork and stakeholder buy-in, reduce the risk of customer resistance, and enable operations to be more efficient in achieving the full value of an initiative.
Assess your agency’s culture to better understand which key stakeholders need to be involved in your data management initiatives and what methods will be effective in implementing them.
Include consistent and ongoing communications tactics for any desired data initiatives to help stakeholders understand why they should adapt to new changes.
Prioritize high-value impact projects, such as those at the management or secretary level, and leverage opportunities for data standardization and integration during planned IT projects.
Execute against your OCM plan, including communicating any upcoming and major milestones of your planned data projects to key stakeholders.
Establish strong metadata tags on data assets helps you find and manage data effectively.
Leverage relationships with data stewards to participate in these foundational data standardization efforts. This activity also supports data cleansing efforts for any forthcoming data projects.
Collaborate with data stewards to identify organizational silos throughout your agency, which can uncover redundancies and overlapping data in systems, tools, and processes.
Assess data tool usage and costs across the agency. Individual agency offices may be applying data analytics into their respective programs; therefore, it is imperative to assess current agency data tools including their licensing costs.
Use your target data architecture to provide the infrastructure to share or migrate data from where it is stored to where it can be analyzed. This data structure should include an infrastructure for a high-functioning analytical environment.
Create a solution that solves the needs of different stakeholders and leverages data as an organizational asset, to discourage agency data silos. An enterprise solution may also lower IT costs.
Implement enterprise data solutions will require gathering data from disparate sources within your agency and making sure you have high quality information data.
Consider data aggregation as a tool in all initiatives and projects. This means you need good extract, transform and load (ETL) capabilities, a necessary process for you to optimize your data for analytics.
Link your data management performance metrics to specific business performance metrics.
Design your data management program to drive or meet business outcomes. This means tracing business outcomes back to supporting analytics output and underlying data.
Implement and maintain data management investments, a key role of the CDO.
Use your budgeting and data resource management process to show how you are prioritizing resources to meet agency needs.
Leverage a data stewardship framework to create a stronger partnership with data stewards and increase their participation in data initiatives for the agency.
Identify ways to get as much done as possible early on, especially in areas of collaboration with data stewards. Holding meetings is not enough. To continually promote participation and buy-in to your data management initiatives, you must demonstrate near-term value.
Federal Data Strategy Data Ethics Framework
GSA guide to Inventory.data.gov
Additional Data Standards from resources.data.gov