Defining and celebrating exceptional federal data practices
The federal government has historically been among the largest collectors of data in the world, with the information it compiles being used by policymakers, private companies, members of the public, the media and others.
Federal agencies collect and distribute data on everything from the weather to the economy to agriculture and the environment. That information is used to forecast the future, address the needs of the public, understand how the country is changing and ultimately improve Americans’ lives.
In 2024, the Partnership for Public Service and USAFacts launched the Federal Data Excellence Program to recognize standout federal data products that empower leaders to make better decisions, enable greater agency accountability and give Americans a better understanding of their government.
The program’s inaugural nominations and winning products consisted of 55 data products providing statistics on agencies, programs and issues across government, covering topics that range from immigration to veterans’ benefits to federal spending. We aim to evaluate year-to-year changes in these data products to assess progress toward FDEP’s objectives.
At the same time, a lot has changed over the past year, with potential implications for federal data. In 2025, the Trump administration reshaped the operations and management of the federal government, leading to staffing and funding changes across many agencies. In addition, the federal government experienced a historic 43-day shutdown in the fall.
Over the past year, federal data collection and dissemination in some domains also transformed. Some new data products, such as the Department of Labor’s Open Data Portal, have aimed to improve public accessibility to government data. Others, such as the National Oceanic and Atmospheric Administration’s “billion-dollar disaster” database and the Department of Agriculture’s Household Food Security Report, have been canceled or stopped collecting data. While there have been some news reports on individual changes, it has been difficult to clearly assess the breadth of change across government data products.
In this context, we examined the current state of FDEP’s inaugural data products, offering a case study in how the accessibility and transparency of data changed in the past year.
Overall, we found that about half of the products that were online in 2024 remained fully functional and had been updated in 2025. The other half missed updates, were not updated at all or experienced reduced functionality. One had notable improvements.
We grouped each product nominated for a 2024 Federal Data Excellence Award into one of three categories:
The vast majority of our 55 nominated products, 84% in total, were expected to be updated, and were still accessible as of December 2025. A much smaller number, 13%, were classified as point-in-time products, and another 4% were taken offline.
Among the products taken offline include the Federal Emergency Management Agency’s dashboard on disaster declarations made by states and counties, along with all of FEMA’s interactive data tools. The dashboard was used to track the areas that declare disasters during extreme weather events or other emergencies.
The Federal Highway Administration’s Inequity Identification Tool, which estimated the cost of air pollution from roads across America and its impact on various demographic groups, was also taken down, likely to comply with executive orders relating to diversity, equity, inclusion and accessibility.
We assessed the functionality of the 46 products that remained online in 2025 and were not point-in-time products. The goal of this assessment was to determine if the datasets were being updated, if they had missed updates or if they had not been updated all year. We also noted if the product had noticeable differences in its functionality since 2024, with the goal of understanding how a data user’s experience may have changed.
About half of the products we reexamined—48%—were online and providing updates as expected. Another 22% missed an update or reduced their update frequency, while 17% were not updated at all in 2025. A total of 11% experienced reduced functionality, while 2% underwent documented upgrades. Of the accessible products, many missed scheduled updates and some experienced a breach in data collection due to the 43-day government shutdown. Still others lacked any updates from 2025 or were discontinued entirely. For example, the National Oceanic and Atmospheric Administration’s Billion-Dollar Weather and Climate Disasters dashboard, which cataloged the impacts of natural disasters and extreme weather on the economy, was discontinued and will no longer be updated.
To take a closer look at what is happening across the federal data community, we also reassessed last year’s winning products using the standards developed by the Partnership and USAFacts. These standards prioritize products’ accessibility, helpfulness, interoperability and timeliness.
We found that the state of these products varied widely. Some experienced minimal changes, while others missed critical updates or had their required data collection canceled.
Despite the shuttering of the U.S. Agency for International Development in 2025, the Foreign Assistance Dashboard, which tracks and details the nation’s aid to other countries and was previously jointly run by USAID and the StateDepartment, remains both online and functional. Its score, based on our Federal Data Excellence Standards, also remains unchanged from 2024.
However, this product was briefly taken offline from Jan. 31-Feb. 3, 2025. This pause occurred while the administration took USAID’s website offline, ended operations at the agency and integrated its remaining programs into the State Department.
The Federal Employee Viewpoint Survey Dashboard, which reports results from the annual Federal Employee Viewpoint Survey, had been administered by the Office of Personnel Management since 2002.
Typically, the Federal Employee Viewpoint Survey is conducted in late spring and results are published in the fall. However, in 2025, the survey was canceled for the first time. As a result, there were no updates, although the dashboard remains online.
When we regraded the Federal Employee Viewpoint Survey Dashboard, it received a lower score than it did in 2024. While the dashboard notes its annual update cadence, it lacks a notice or message on the site’s homepage or associated pages about the survey’s cancellation.
Additionally, 13 questions related to diversity, equity, inclusion and accessibility, which were added to the survey in 2022, are no longer on the dashboard. Those questions were redacted to comply with executive orders related to gender and diversity, equity and inclusion programs.
Despite significant cuts and staff reductions at the Department of Health and Human Services, the National Survey on Drug Use and Health State Estimates Dashboard is both online and functional, without major changes.
On July 28, 2025, the survey’s annual national report was published, highlighting key findings from the 2024 survey. However, the dashboard’s data is based on two-year estimates, so updates with 2024-2025 data are not expected until 2026. Given the overall status of this product, its grade from 2024 remained the same.
Although the survey remains online and functional, its future is uncertain. The team that orchestrated the survey was laid off in April, making it unclear whether data collection will continue going forward. Additionally, the 2024 survey report omitted information about drug use based on race and ethnicity, data that was included in previous years’ surveys.
The Higher Education Research and Development Survey Dashboard, created by the National Center for Science and Engineering Statistics at the National Science Foundation, tracks how much colleges and universities spend annually on research and development.
The dashboard received the same score as in 2024, indicating that there were no major changes to its accessibility, helpfulness, interoperability or timeliness.
The survey website indicated that new data would be released in November 2025. However, it was only updated in late December, and there were no notices posted in November about an updated release date.
This evaluation of government data products highlights some of the vulnerabilities and challenges facing federal data. While many of the nominated and winning products remain accessible and usable, many others have gone without updates, missed deadlines or become less functional, ultimately making them less useful to policymakers and others who make critical decisions that affect the public.
The federal government is the only institution with the resources and capability to collect and analyze data on a multitude of topics and at great scale. This information is used to shape policy, inform the public about issues that impact them and better understand how the country is changing. Ensuring consistent and timely data publication — and being transparent about when and why disruptions occur — builds trust in government and ultimately serves the public interest.
Products were reviewed and regraded from late October to late December 2025.
To regrade 2024’s winners, the winning products were assessed using the same version of the Federal Data Excellence Standards applied previously. Each product’s grade was based on 22 key weighted metrics that evaluate the product’s helpfulness to the user, data accessibility and timeliness, and interoperability.
Products were either tagged as online, offline or point-in-time, based on whether they were able to be viewed on the web and whether they claimed to include updated datasets or those from a particular time frame.
Product functionality was based on whether the product had been updated in 2025, and if so, whether it had been updated according to the cadence laid out. If a product was updated, it was marked functional. If it was not, it was either marked as having missed updates or not having been updated at all.
Finally, if a product had one or more functions clearly missing or broken, it was marked as having impaired functionality. If it was both updated and had received documented improvements or upgrades, it was marked as improved.
