A savings claim needs an evidence trail.

What is included?

Eight selected topics from GAO’s 2026 duplication, overlap and fragmentation report, joined to one specific underlying recommendation or congressional matter each. This is a curated pilot, not all open recommendations, all federal waste, a fraud register, or an exhaustive savings estimate. GAO provides the broader official tracker and downloadable data.

The original contribution is the plain-language join: responsible institution, exact selected action, GAO status, latest selected comment date, topic-level benefit context, evidence-to-close summary, and an original evaluation checklist. GAO owns the official status assessment. Our proposed tests are not official findings or instructions from GAO.

One record has three different clocks.

1. Publication and public release: when the underlying report became available. These can be different dates.
2. Latest selected comment: the date mentioned in the status commentary for that action—not necessarily the whole report’s newest comment.
3. Source consulted: when we checked the source. A retrieval today does not turn an older comment into a newly confirmed implementation status.

The records show the status in the consulted official page. They do not claim independent verification of every agency development since the selected comment. An agency’s statement that it implemented an action is not GAO closure. A planned date passing does not prove either success or wrongdoing.

Why there is no grand total

The annual report cautions that estimates have different sources, periods, assumptions and methodologies. Some describe cost reductions, some revenue, some historical policy effects, and some unquantified service improvements. Implementation costs, interactions, partial action, and overlap can change what is realized. Adding them would produce a misleading answer.

A topic-level estimate is not a price tag solely on our selected recommendation. The desk does not reproduce CBO, JCT or agency models; the attribution is “as cited by GAO.” Exact baseline dates or model assumptions not supplied in the annual table remain unspecified. We do not invent an annual rate or a confidence score to fill those gaps.

A financial opportunity also does not create a funding entitlement, IOU, or automatic allocation to America.co projects. Any actual reallocation would require its own lawful authority and budget process.

How to audit the join

1. Open the underlying report using the stable GAO report number.
2. Locate the selected recommendation number or the exact congressional matter. Our colon-suffixed join keys are desk identifiers, not GAO database IDs.
3. Read that action’s status and comment—not a neighboring recommendation’s status.
4. Open the annual report topic and estimate notes. Keep its scope distinct from the selected action.
5. Check whether the claim is a potential estimate, measured result, historical estimate or unquantified benefit.
6. Ask what evidence is needed, who can provide it, and what would change the interpretation.

JSON records and CSV export retain source links, selected actions, consultation date, estimate scope, periods and limits. Blank and editable checklists are research aids, not official requests. Downloaded personal worksheets are not transmitted or automatically stored.

Attribution, rights and privacy

Compiled by the America.co public-resource desk, with AI assistance and linked primary sources. No individual expert authorship or expert review is fabricated. America.co is independent of GAO and the institutions named. This is public-interest explanation, not individual medical, tax, legal, repayment, or engineering advice.

The annual report’s rights notice identifies it as a U.S. government work and warns that some images or other material may have separate copyrights. We use original summaries and our own branded artwork, not copied GAO graphics. No private addresses, personal voter or borrower records, or victim details are part of this dataset.

We reviewed the CC0 Inspectors General project for stable report identifiers and provenance patterns. Its historic scrapers are not imported, and its coverage is not claimed as ours. The desk reuses America.co’s tested static publishing and private worksheet components.

Refresh and correction policy

Recheck selected sources at least monthly and after a relevant status change. The release checker verifies the selected row, status and comment date and records source hashes. A mismatch requires editorial review; it does not silently rewrite public conclusions. No continuous or automatic update service is claimed.

Send a correction through the contact page, including the public passage, exact selected action, source URL, and date. Omit personal identifiers. Material corrections update the relevant record and its change note. If evidence is unavailable, the gap remains visible.

Initial release: October 5, 2026; eight curated records, comparison, exports, local-only checklists, and this methodology. No traffic growth, indexing success, implementation result, recovered funds or new backlinks are inferred from publication.