For the Government Accountability Office, inspectors general and oversight bodies

Oversight finds the pattern. Evidence makes it stick.

Auditors already know where the money goes wrong. What they rarely have is evidence that covers the whole population, survives challenge and does not depend on the agency under review.

Open the GAO demoRequest a briefing
Chapter IThe scale

Improper payments are reported. Fraud is estimated. Both are larger than anyone can sample.

Federal agencies reported about $186 billion in improper payments for fiscal year 2025, and nearly $2.8 trillion since 2003. The reported figures cover only a subset of programs.

GAO's own estimate puts annual fraud losses at $233 billion to $521 billion, between three and seven percent of federal obligations.

GAO, fiscal year 2025 $186B in improper payments reported by federal agencies, up $24 billion from the year before.
$233B to $521Bestimated annual fraud losses, FY2018 to FY2022
$2.8Tcumulative improper payments, FY2003 to FY2024
Sources: GAO, April 2026; GAO-24-105833; GAO-25-108172.
Chapter IIWhat an auditor needs

Four things a sample cannot give you.

The whole population

Every payment tested, not a sample extrapolated, so findings stand on their own.

Reproducibility

The same rules, the same data and the same result, every time anyone reruns the test.

Chain of custody

Proof of where the data came from and that nothing changed between extraction and report.

Independence

A method the agency under review does not own and cannot quietly adjust.

Chapter IIIWhat we provide

Workpapers that defend themselves.

We test the full payment population against the program's own rules, link related parties, and seal every result with the rule, the record and the time. Root causes are measured rather than estimated, which is what the Payment Integrity Information Act asks agencies to report.

  • Full-population testingEVERY PAYMENT
  • Root-cause evidence for improper payment reportingMEASURED
  • Sealed workpapers with chain of custodySEALED
  • Self-authenticating exhibits under FRE 902(14)CREB
  • No contingency fee, no stake in the outcomeINDEPENDENT
Chapter IVIndependence

We take no share of recoveries. Our findings have no reason to be large, only to be right.

We own no health plan, run no provider network and hold no contract that pays us by the dollar found. That independence is part of the evidence.

Bring us an audit in progress. We will test the whole population.

Share the program and the question. We will propose a full-population test and show what the sealed workpaper looks like.

Open the GAO demoRequest a briefing