The Office of Management and Budget approved and released the Federal Data Strategy’s first-year action plan Dec. 23, marking a milestone in how agencies will implement the strategy within their organizations and in collaboration across the government in 2020.
Now that the action plan is out, agency co-leads of the strategy and agency chief data officers who helped develop and are implementing the plan provided insight about how it will help agencies navigate new data-driven requirements from policies such as the Data Accountability and Transparency Act and the Foundations for Evidence-Based Policymaking Act, or Evidence Act.
What’s in the First-Year Action Plan?
The Federal Data Strategy Development Team released a draft of the action plan earlier in June and from there accepted public feedback to review and apply to the final action plan. Initially supposed to be released in September, the plan was delayed pending approval from OMB, according to Data Strategy Development Team Co-Lead and Small Business Administration CIO Maria Roat.
The final plan includes feedback from the draft action plan, which included 16 actions. The finalized version has 20 actions that agencies will have to complete throughout 2020.
The final actions are divvied into agency-specific and cross-agency categories. Of the 20 actions, there are six each agency will have to complete on their own, four that agency communities of practice and OMB will oversee, and 10 shared, interagency actions each helmed by an individual agency.
At the individual agency level, the final action plan calls for each agency to:
- Identify data needs to answer priority agency questions by September 2020
- Institutionalize agency data governance by first publishing data governance body materials documentation on their data sites by January 2020 then beginning enterprise-wide data governance activities by November 2020
- Assess data and related infrastructure maturity by May 2020
- Identify opportunities to increase staff data skills by July 2020
- Identify priority data assets for agency open data plans by January 2020
- Publish and update data inventories by July 2020
As each agency strives to complete these six actions over the next year, there will also be collaborative communities of practice actions, which largely touch on many preexisting initiatives, such as creating structure for cooperative data leadership and standardization. More specifically, the action plan states that these community of practices should:
- Launch a federal CDO Council by January 2020 under the leadership of OMB
- Improve data and model resources for artificial intelligence research and development by February 2020
- Improve financial management data standards a key goal of the President’s Management Agenda by August 2020
- Have the Federal Geographic Data Committee and OMB Federal Data Policy Committee integrate geospatial data practices into the federal data enterprise by December 2020
The final 10 actions in the first-year action plan are “shared solution actions,” or government-wide data services each led by a singular agency or small group of agencies. These final items call for:
- The General Services Administration, OMB and the Office of Government Information Services of the National Archives to develop a repository of federal enterprise data resources — such as case studies, the data.gov schema, playbooks, requirements, tools and standards — by December 2020
- OMB to establish a Federal Data Policy Committee by January 2020
- GSA to develop a curated data skills catalog “of learning opportunities to help agencies develop competencies for managing data as a strategic asset” by November 2020
- GSA to develop a data ethics framework “to help agencies systematically identify and assess the potential benefits and risks associated with data they acquire, manage and use” by December 2020
- U.S. Census Bureau Federal Strategical Research Data Center Program Management Office to pilot a one-stop standard research application by March 2020
- National Center for Education Statistics to pilot an automated tool for information collection review that supports inventory creation and updates by July 2020
- GSA to develop and pilot an enhanced data management tool for agencies by September 2020
- Federal Committee on Statistical Methodology to develop data-quality measuring and reporting guidance by December 2020
- GSA to develop a data standards repository by December 2020 in collaboration with OMB and the National Institute of Standards and Technology
With the expanded list of actions, which were included for clarity purposes, the overall content and activities in the final plan are almost identical to those in the draft.
“Many commenters provided feedback on the 16 actions included in the draft action plan, identifying opportunities for clarification and suggesting additional Actions to address identified gaps,” the document says. “The final action plan offers additional context, clarity and detail.”
Since June, 140 unique commentators from government agencies, academia and industry provided feedback on the data strategy website, and a collective 66 individuals offered comment at two forums in June and July.
Roat said that that the transparency that OMB and the data strategy development team maintained throughout the drafting and editing of the action plan minimized surprises for agencies or industry partners and enabled positive collaboration in finalizing the action plan.
Roat added that because of the nature of the project, she or the groups she collaborated with expected few changes in the final plan from the draft released in June.
“When you look at the federal government and how it manages data, getting us to be more consistent, making sure we’re protecting security, privacy, confidentiality — being able to do all of those things but being consistent across the federal government is really where we need to drive, and that’s where the strategy is going,” Roat said. The action plan "is a big step for the federal government, a positive one.”
Department of Commerce Chief Data Officer Ed Kearns added that the transparency behind the process of the action plan, and the Federal Data Strategy’s tenets, help agencies meet and even exceed the requirements of the DATA Act and Evidence Act, which require greater transparency in federal spending information and for agencies to increase data-driven decision-making.
“The compliance exercise is outlined by the Evidence Act,” Kearns said, adding that the Federal Data Strategy and action plan are “more aspirational" than the Evidence Act. "The target for the data strategy is definitely higher. Hopefully, as we implement this strategy, we will find ourselves needing to find the Evidence Act on our way to achieving the data strategy goals.”
What does federal leadership think?
Although the White House in January signed the Evidence Act — requiring all agencies to appoint CDOs to help agencies leverage data in their activities — the act did not provide guidance on where CDOs belong within their organizations or how to integrate data-driven governance into the business of government.
For agency CDOs, the ambiguous nature of their roles has created several challenges. Kearns and SBA CDO Emily Knickerbocker both said that some agencies are just coming to understand the value of data in their activities and have CIOs doubling as CDOs. Others are further along in tying data to mission, such as Commerce, which houses data-driven agencies like the Census Bureau and the Patents and Trademark Office, or have had well-established CDOs for years, such as Knickerbocker, who has been SBA’s CDO since 2017.
Despite these challenges, the data strategy’s action plan — which heavily emphasizes collaboration and transparency — have provided the guidance and tools agency CDOs and data scientists will need to get on the same page in leveraging data as an asset, according to Roat, Knickerbocker, Kearns and GSA CDO Kris Rowley.
“When you talk about the overarching governance — ethical governance, conscious design, learning culture, the security and the privacy, and all of those elements — to come together as a federal government and have consolidated principles for leveraging data as an asset, I think, was big to the federal government to come to an agreement on what those principles would be,” Roat said.
There were several elements of the action plan that CDOs Knickerbocker, Kearns and Rowley said they found particularly compelling and helpful to promote the collaboration and take the reins on data management with the increase in quantity and availability of data.
Rowley said that he looked forward to the formation of a Federal CDO Council, which he said will help different agencies and their CDOs get on the same page in their leadership functions, thereby enabling greater interagency cooperation.
“I would recommend that we primarily focus on the governance structure each agency is trying to roll out to make sure that we have some level of consistency around the authorities and responsibilities of CDOs,” Rowley said of the to-be CDO Council. “I know that it won’t be totally consistent because CDOs are being placed in different organizations across the government, but if we can establish some consistency of governance models, it will help us as we try to do interagency collaboration.”
Rowley added that as the data strategy brings CDOs together and encourages cross-agency collaboration, he envisions that agencies will also start to better underscore data quality across agencies.
“Bringing together agencies under a common Federal Data Strategy will drive greater emphasis on data quality and making sure that when we’re doing analysis that we’re using data to the greatest extent possible rather than rebuilding or recollecting data that we already have,” he said.
Knickerbocker and Kearns also see the action plan’s emphasis on a learning culture that cultivates staff data skills.
“There’s a lot of competition across all organizations to find people that have the right skills to help them analyze and understand the new and large amount of data that people are trying to leverage as an asset,” Knickerbocker said of current personnel challenges.
Kearns added that the government currently has many employees who could be considered data scientists, such as meteorologists or economists, and that developing a learning strategy to get current personnel who handle data trained to use data skills that the government will need in the future is also an element of the action plan he finds critical.
The action plan’s call to create a data ethics framework is crucial, Kearns said, as it will create a foundation to how government decides to utilize data.
Of all the items, however, identifying and forming open data plans and the collaboration and transparency required behind those processes, are the most exciting components of the action plan.
“The sharing of our data, the sharing of our software, sharing of our documentation — in a way that makes them even more valuable as an asset to the public, to industry, to academia, researchers and other federal agencies,” Kearns said. “Being able to share our data effectively is one of the biggest challenges that we’ve had across different communities and agencies, and I think this action plan is going to help us bring that to the next level.”
“A key component of the data strategy is promoting transparency,” Roat added. “I’ve always been really big on collaborating, sharing data.”
The action plan’s call to open and collaboratively leverage agencies’ wells of data will hopefully get agencies who are in different stages in the Federal Data Strategy’s implementation on the same page, Knickerbocker added. As agencies continue to collaborate under the guidance of the action plan, they can start to tackle individual data-driven projects within their own organizations.
How will the action plan impact individual agencies?
In addition to the broader benefits the Federal Data Strategy and its action plan will bring to federal agencies, the action plan will bring about specific data-driven projects.
Kearns touched upon data-backed projects at Commerce, which has agencies that maintain some of the biggest wells of data across federal agencies. The federal plan will preempt Commerce’s data strategy, which will set the guidelines for how the bureaus within Commerce will use and share data across the department.
“We’re not just independent bureaus and agencies,” Kearns said. “We do have important cross connections that, if we utilize them well, will benefit the American taxpayer by providing either our services for efficiency or providing data in such a way that they can be used more easily together.”
If Commerce can take census, economic and environmental data from its different bureaus and properly bring them together, not only agencies but also researcher and industry partners can conduct meaningful work with that information.
“It’s a very powerful tool for industry and researchers to start to understand how the economy is being influenced by things like climate change and by weather ... and respond to population changes and economic changes,” Kearns explained. “I think it’s going to help the Department of Commerce figure out how it’s going to lean into some of these partnerships with industry that I think are very valuable.”
GSA also has its own data-driven goals that the data strategy's collaborative aspects can help with. An initiative within GSA to bring its customer agency data together will allow the agency to interact with them and show the current state of their deliverables, Rowley said. The data-sharing and governance structure that the data strategy emphasizes will help GSA in this effort.
“Some of the initiatives are related to having more of a centralized investment strategy from a technology perspective of, 'How do we take a look at our enterprise infrastructure to better support these types of more centralized analytic capabilities?'” Rowley said. “I see the action plan and Evidence [Act] combined really helping to drive home a stronger governance structure, which in turn should help us make decisions on making the right type of investments.”
While Rowley sees how GSA can help its own mission and back-office functions for other agencies with the data strategy and action plan, the plan has helped SBA’s leadership continue to reinforce data-driven change within the organization — both for the agency itself and the small businesses it supports.
SBA has already started providing better data analytics tools for its personnel the past couple of years, Roat said. The agency also plans to develop a “360 lifecycle” of a customer — which will require much better understanding of data and how small businesses interact with SBA for aid.
“A small business may start out with counseling and then apply for a microloan, and then apply for a certification from SBA. The ability to have a 360-degree-view of the customer means having access to the common data across the agency and then using that data to inform decision-making and better serve small businesses,” Roat said.
With the expanded view, SBA will be able to be more predictive with its data so that it can accelerate how the agency helps its customers. As SBA comes to better understand its data, the agency can better plan its field offices and their resources across the country as they engage with small businesses, Roat added.
Not only will getting a better understanding of SBA’s data help the agency in its activities, but also it can help the small businesses that want to leverage the data that SBA would make more open under the Federal Data Strategy.
“Making data open by default is beneficial to industry partners and small businesses because it really democratizes the data,” Knickerbocker said. “It’s not just the people who are familiar with the data and put in a [Freedom of Information Act] request who are able to get access to the data. It’s everyone getting access at the same time.”
As excited as agency leadership may be for the opportunities and guidance within the Federal Data Strategy, the first-year action plan only scratches the surface in the long-term mission ahead.
Although much of the work will occur over the next year, executing the mission of the data strategy will take longer.
“It’s going to take multiple years,” Roat said. “There’s a long-term maturity to get the full value of federal data. It’s not just about the data itself, but about delivering a more consistent approach to federal data stewardship, use and access.”
In total, the Federal Data Strategy site states that the data strategy will take place over a 10-year period with a new action plan release each year to keep agencies on course throughout the decade. Future action plans will build on those of previous years.
Although 2020 and the years to follow may be “a little daunting,” as Kearns framed it, CDOs and other federal IT leadership across agencies realize the importance of creating a strategy and plan to maximize the possibility of government data. Kearns said that the data strategy has given agencies the opportunity to be aspirational in leveraging data as an asset, and the first-year action plan, in many ways, is a springboard in allowing agencies to shoot higher in its data-driven goals.