Photo: Wikimedia Commons, public domain, US Government Accountability Office. Image source (opens in a new tab).

Trump's Budget Office Wants to Retire the Federal IT Dashboard. Auditors Found 24 of 53 Major Projects Looked Riskier Than Reported.

Factrail investigative analysis | Washington | 2 November 2026 | Original documentary review; independent editorial and legal QA pending.

The finding. The Trump administration's Office of Management and Budget (OMB) has proposed retiring the existing public Federal IT Dashboard in favor of a streamlined replacement, while an October 7 audit finds that the government's warning signals about major technology projects are frequently too reassuring. In a deliberately selected sample of 53 investments at 12 agencies, the Government Accountability Office (GAO) assessed greater risk than the agency chief information officers reported in 24 cases; 21 ratings were not refreshed on time under the agencies' own rules. OMB had announced no replacement timetable to GAO. That does not prove $100 billion was wasted, that contractors manipulated scores, or that every disagreement was misconduct. It establishes a narrower, consequential accountability problem: the administration wants to change the public instrument just as auditors show why its underlying inputs and statutory review process need repair. GAO, October 7, 2026 (opens in a new tab) and Federal News Network, May 12, 2026 (opens in a new tab).

What this investigation adds

The public debate tends to present a binary choice: keep an old, labor-intensive website or replace it with something modern. The more useful test is institutional. A dashboard is only the visible last step in a chain: program managers record risks; agency chief information officers translate them into ratings; OMB sets reporting rules and intervenes in persistently troubled projects; GSA runs the public interface; Congress and taxpayers examine the result. This article reconstructs that chain against the GAO's project-level examples, the administration's stated case for streamlining, the 2027 GSA budget request and the federal technology leadership transition. The documentary question is whether a replacement can preserve legally required public warning signals and make them more reliable, not whether a particular website deserves to survive.

The original calculation from the GAO comparison is straightforward: 24 divided by 53 is 45.3%, while 21 divided by 53 is 39.6%. Those are shares of a targeted audit sample, not estimates for the whole federal portfolio. GAO selected projects from budget data with at least $35 million planned in fiscal 2025 development, modernization and enhancement spending, then excluded cases without adequate risk documentation. The $100 billion-plus annual federal IT-and-cyber spending figure is a government-wide scale marker, not the value of the 53 sampled projects or a proven loss. GAO full methodology (opens in a new tab).

The documentary timeline: a law, a refresh, a retreat

2009: OMB launched the Federal IT Dashboard to expose information about large federal technology investments. The underlying problem was not merely a lack of attractive charts. Large IT procurements can continue spending while schedules slip, requirements change and technical weaknesses accumulate. Public visibility was meant to give agency leaders, legislators and outsiders a common signal.

December 2014: Congress enacted the Federal Information Technology Acquisition Reform Act (FITARA) provisions as part of Public Law 113-291. GAO explains that the framework requires public information on major investments and creates review obligations for projects rated high risk for four consecutive quarters. This is the legal distinction OMB cannot resolve merely by announcing a different website: statutory disclosure and oversight duties outlive the interface. GAO's statutory and portfolio-management review (opens in a new tab).

November 2024: GAO reported that OMB and agencies had not fully performed the high-risk review machinery. It found 17 major investments at eight agencies that had remained high risk for four consecutive quarters without any of those eight agencies fully meeting the associated FITARA requirements. In its examination of 27 review sessions, the Federal CIO had not been consulted. These were prior audit findings, not evidence that the same 17 investments remained in that condition in November 2026.

August 2025: OMB updated capital-planning guidance, including a requirement that chief information officer risk ratings be refreshed at least quarterly. The new instruction made the cadence more explicit, but a written cadence cannot itself correct an inaccurate assessment or force a late submission.

April 2026: Then-federal CIO Gregory Barbaccia published a letter saying the dashboard was costly and inefficient and that agencies would shift to a more streamlined state focused on legally required data. His promise was that the public would continue to receive such data. The exact day of the April announcement is not established in the documents reviewed here; it should not be invented. The contemporaneous Federal News Network account (opens in a new tab) reproduces the substance of his position.

April 2026 budget request: GSA's fiscal 2027 congressional justification requested resources to move responsibility for the dashboard from its Office of Government-wide Policy to Technology Transformation Services. A separate budget-neutral transfer of $12.1 million and 12 full-time-equivalent positions covered several programs, including the dashboard, accessibility and ROCIS/Reginfo. It is not a $12.1 million dashboard-only contract or an awarded saving. GSA FY2027 justification (opens in a new tab).

August–October 2026: OMB confirmed that Barbaccia would leave federal service at the end of August. On September 30, specialist outlets reported that EPA technology chief Carter Farmer was expected to succeed him, but the reporting available here did not independently establish the formal appointment date. On October 7 GAO released its 73-page risk-rating audit, with 17 recommendations to nine agencies. This sequence means the public promise to modernize and the documented weakness in oversight span a leadership handover, not one official's uninterrupted tenure. Departure report (opens in a new tab); successor reporting (opens in a new tab); GAO audit (opens in a new tab).

The risk numbers, without the misleading denominator

GAO compared April 2025 agency ratings with its own standardized assessment based on risk registers, review-board material and other performance documentation. Its judgments agreed with agency CIO ratings for 27 investments, indicated greater risk for 24, and indicated less risk for two. That last category matters: the auditors were not mechanically assigning higher danger to every project. They identified some cases in which agencies were more cautious than the GAO framework.

The disparities are not evenly spread. For the six Department of Veterans Affairs projects in the sample, GAO found more risk in all six. For five of eight Transportation projects, GAO found more risk. Treasury had three of four and HHS four of nine. Agriculture's five selected ratings all matched. Those are descriptive findings for the selected projects, not a league table of overall agency competence.

Project names turn a statistical dispute into a testable question. GSA's Login.gov carried a medium agency rating but a high GAO assessment. GAO pointed to unresolved technical reliability and performance issues and incomplete data-protection implementation applicable to March 2025. GSA argued that its own high-risk threshold required persistent failures on metrics, costs or schedules that it did not see. Treasury's Case Management project was moderately low according to the CIO and high according to GAO. VA's Health Management Platform was also moderately low versus high; its Benefits Payment was low versus moderately high. Transportation's Automatic Dependent Surveillance–Broadcast and Enterprise Information Display System were moderately low versus high. These are names and relative ratings from GAO's table, not findings that those services failed, leaked personal data or caused harm. GAO project-level table (opens in a new tab).

The methodological dispute is substantive. CIOs often incorporate contract management, workforce, security, requirements and performance, not simply the numerical severity of entries in a risk register. GAO explicitly says its standardized method was not intended to prescribe every agency's methodology. The accountability issue is whether broader composite scores obscure material threats that should be visible to decision-makers. That question is answerable through contemporaneous registers, rating rationales and escalation records, not by accusing every CIO who disagrees with GAO of dishonesty.

The second defect: stale warnings

The risk mismatch is one problem; timeliness is another. GAO found 21 of 53 selected ratings were late relative to the relevant agencies' own procedures. Its table attributes nine to HHS, five to Commerce, four to Homeland Security, and one each to Education, Transportation and GSA. The snapshot was April 2025, so it must not be represented as a real-time count for November 2026.

HHS officials described staffing reductions in April 2025 while personnel were compiling ratings. The department also allowed some components to report semiannually rather than quarterly. Education's own process was semiannual. Both conflicted with the later OMB minimum-quarterly direction. Agencies can have genuine capacity problems, but a reporting schedule that is slower than the required cadence defeats the function of a warning light: a red flag seen months later cannot trigger timely corrective action. GAO findings and agency explanations (opens in a new tab).

The distinction between late and understated ratings also prevents double counting. Some of the 21 stale ratings may overlap with the 24 for which GAO found greater risk. Adding them together as 45 separate failed investments would be mathematically wrong. The strongest defensible statement is that the sample showed both kinds of weakness, sometimes potentially affecting the same investment.

Who actually holds the power, and who can block a fix?

The President and OMB director. The Trump administration sets executive-branch technology priorities; OMB under Director Russell Vought controls the government-wide policy and budget framework. But neither a presidential preference for lower reporting burden nor an OMB policy letter repeals congressional reporting obligations. OMB can redesign the data collection mechanism, change administrative guidance within legal limits, and require agency reporting. Congress can demand the legally specified information and scrutinize any shortfall.

The Federal CIO and agency CIOs. The federal technology office sets expectations and participates in reviews of persistently high-risk investments. Agency CIOs exercise the first-line judgment over individual ratings and supporting evidence. Barbaccia made the April announcement; later staffing and succession decisions matter because the remedy requires a continuing office, not a press statement. GAO said its earlier recommendation concerning federal CIO participation in high-risk reviews remained unimplemented as of October 2026. This is a reported oversight gap, not a court finding that a named official deliberately violated the law.

GSA. GSA operates the public dashboard and can change its engineering, usability and data delivery. It does not independently determine the truth of every agency risk score. GSA's 2022 modernization announcement described improved search, visualizations and public API access. The 2027 budget request points toward a different organizational home in Technology Transformation Services. Neither proves a replacement is operating or that data quality has been repaired. GSA's 2022 launch (opens in a new tab); GSA budget request (opens in a new tab).

Congress and GAO. The House Oversight and Government Reform Committee's chair James Comer and ranking member Robert Garcia received the October GAO report. The committee can request records, hold hearings and assess whether the new system meets the statute. GAO can make and track recommendations; it cannot itself compel an agency to correct a rating or replace a project manager. The operational veto over a particular contract, appropriation or program remains with the relevant lawful decision-maker.

Money: who receives it and who might benefit?

The immediate financial beneficiaries of federal IT spending are vendors that deliver technology, integration, maintenance and consulting, as well as government organizations receiving services. A rating that is too reassuring could, in principle, reduce pressure to revise a troubled procurement or contract. But the audit does not show which contractor, if any, obtained an extra dollar because of a particular rating. There is no evidentiary basis here for assigning a fraudulent profit to a vendor or a personal financial interest to Vought, Barbaccia, Farmer or congressional officials.

One named supplier does appear in independent reporting on the dashboard itself. Federal News Network reported on May 12 that GSA's dashboard contract with Booz Allen Hamilton was about $3 million annually and that GSA had spent approximately $25.4 million since a 2021 award, citing USASpending.gov. We did not independently reconcile individual federal award modifications, obligations and outlays from the primary procurement ledger for this draft. Those figures therefore remain attributed reporting, not a Factrail-certified cash-flow reconstruction. Booz Allen is an identified dashboard contractor, not an identified beneficiary of inaccurate risk ratings. Federal News Network contract reporting (opens in a new tab).

The administration's best fiscal case is that manually assembled, self-reported information can be expensive and stale before publication. A more automated and narrower reporting system could reduce labor and platform costs. The test is a before-and-after accounting: actual operating and transition costs, which data fields disappear, what risk evidence remains public, and whether agency staff spend less time submitting while oversight becomes more effective. Savings cannot be asserted solely because a website is being retired. Equally, a contractor payment cannot be labeled waste without comparing the services delivered, the procurement terms and credible alternatives.

GSA's requested $12.1 million multi-program budget-neutral transfer is a proposed allocation change, not evidence of a windfall to Technology Transformation Services or a payment to Booz Allen. The distinction between an agency budget request, a contract ceiling, an obligation and an outlay is essential to this story.

The strongest counter-case is in the auditors' own report

The old dashboard was not a model of flawless public administration. Seven of the 12 agencies reported usability, data-quality or submission difficulties. Users cited unclear error messages, inadequate bureau-level analysis, missing rating history and publication delays. OMB said its new approach would focus on data actually required by statute and lower agency burden. A leaner tool might be a better oversight tool if it retains meaningful history, standardized ratings and accessible downloads.

The GAO also found evidence against a simplistic story of systemic neglect by GSA. In a survey of 37 agency users in 2024, 26 were somewhat or very satisfied overall; 28 were satisfied with the effort required for tasks. GAO credited GSA with responding to user feedback and found that reported challenges did not prevent core functionality. The problems were not all attributable to the website's design. GAO survey and conclusions (opens in a new tab).

Agencies disputed GAO's rating methodology. Defense argued that the auditors' alternative score did not capture the full method used for its Army personnel-and-pay investment. Commerce defended its Trademarks Product Line assessment and explained a delayed update as an interagency agreement issue. DHS maintained that its existing composite Program Health Assessment was sufficiently risk-informed, while agreeing to improve timeliness. Treasury cautioned that different frameworks do not necessarily mean the agency's judgments were inaccurate. Those responses must be read alongside the auditor's explanation that it examined risk registers and other contemporaneous evidence. GAO agency responses (opens in a new tab).

Five agencies agreed with all recommendations directed at them; DHS agreed with one and disagreed with another; Commerce and Defense disagreed; Treasury neither agreed nor disagreed. The GAO's 17 recommendations are therefore an unresolved governance agenda, not 17 adjudicated violations. The strongest version of the administration's case is not that GAO found nothing wrong. It is that the current collection system imposes cost without reliably producing current information, so replacing it and strengthening CIO responsibility could improve both economy and accountability. That proposition deserves a measurable trial.

The public consequences are concrete, but not yet quantified losses

A federal technology risk rating is an early warning about an investment's ability to deliver its goals. The sampled projects support functions such as veterans' benefits, tax administration, student-loan servicing, airport surveillance and federal identity access. A misleading rating could delay escalation, leaving service users exposed to future outages, delays, privacy problems or unnecessary spending. Could is doing important work: GAO's comparison did not establish that each project suffered those outcomes, or quantify a causal dollar loss from each mismatch.

The public's second exposure is informational. If the old dashboard disappears before a successor preserves historical ratings, definitions and download access, Congress, watchdogs, reporters and citizens may lose the ability to reconstruct when officials knew a project was deteriorating. A new interface could instead make those histories more usable. The critical question is whether the government retains an audit trail that cannot be silently rewritten after an embarrassing failure.

What would change the verdict?

A credible remedy would include a dated, publicly inspectable replacement specification; continued access to statutory project-level risk information; archived historical ratings and revision dates; transparent explanation of changed scoring methods; and documented Federal CIO participation when the four-quarter high-risk review threshold is met. Agencies could then show their risk registers, escalation decisions and corrective actions without disclosing genuinely protected security details. GSA could publish the costs of the old and new services using comparable accounting periods.

Conversely, evidence that the successor drops essential risk histories, that ratings remain stale, or that high-risk projects avoid mandatory reviews would strengthen the conclusion that streamlining weakened oversight. Evidence of lower verified operating costs, more timely reporting, a usable public history and completed statutory reviews would weaken that concern. These are falsifiable tests rather than a predetermined accusation.

There is no documentary basis in the sources reviewed to say Trump personally ordered a particular investment rated green, that Booz Allen or another vendor bought influence over ratings, or that any named politician profited from the dashboard's planned retirement. The documents do support criticism of a governance decision made before the public replacement timetable and risk-reporting repair were demonstrated.

Documentary source register and limitations

Primary evidence: GAO's October 7, 2026 full audit, including methodology, project tables and agency replies (opens in a new tab); GAO's November 2024 FITARA oversight review (opens in a new tab); GSA's fiscal 2027 congressional justification (opens in a new tab); GSA's March 2022 modernization release (opens in a new tab); GSA's April 2026 data-transparency explanation (opens in a new tab). Independent reporting: Federal News Network on the April OMB announcement and dashboard contractor (opens in a new tab); MeriTalk on the October audit (opens in a new tab); Federal News Network on Barbaccia's departure (opens in a new tab); FedScoop on the reported successor (opens in a new tab).

This is an original synthesis of public records and independent reporting, not an exclusive interview or leaked-document investigation. The exact April announcement day, a finalized replacement timetable, full vendor payment ledger, subsequent CIO appointment instrument and up-to-date project-level corrections were not independently established. OMB did not provide comments to GAO on the October report; that does not mean OMB refused a direct Factrail request, because no such request was sent. No legal or editorial approval, verified Factrail internal-link target, image license or production publication is asserted.

Related Factrail context

Broader contextual entries include Donald Trump (opens in a new tab), the White House (opens in a new tab), and democratic resilience (opens in a new tab). These background links do not establish involvement in any particular IT investment.