The Justice Department has posted the form it now wants from the fastest-growing class of False Claims Act whistleblower: the data miner who never worked for the defendant. The FOCUS Initiative White Paper Questionnaire, linked from the Civil Division's announcement page, which was updated July 17, 2026, asks anyone pitching a statistically derived fraud theory to answer seven questions in no more than 10 pages total, name the data sources behind the model, disclose every qui tam case they have filed before, and explain why their single best example is fraud rather than an administrative overpayment.

Background

FOCUS stands for Fraud Oversight through Careful Use of Statistics. The Civil Division launched it on April 30, 2026, under Press Release Number 26-423, and framed it as a response to a structural shift in who brings False Claims Act cases. The division said it has experienced a rapid increase in the number of qui tam complaints filed in recent years, and that much of that surge has been driven by companies or individuals who analyze publicly available government data for potential signals of fraud, rather than the insiders who have traditionally served as relators.

That shift matters because the two kinds of relators arrive with very different evidence. An insider brings emails, invoices, and firsthand knowledge of what a contractor told the contracting officer. A data miner brings a pattern: a billing rate that sits well above peers, a labor category that appears too often, a claim volume that spikes after a rule change. The pattern may be a genuine fraud signal. It may also be a coding convention, a population difference, or a contract modification the model never saw.

Assistant Attorney General Brett A. Shumate of the Civil Division said sophisticated data analytics have become an increasingly important means of identifying fraud trends and uncovering patterns of misconduct across federal programs. Deputy Assistant Attorney General Brenna E. Jenny set the bar for participants: they should be prepared to explain what differentiates their approach, how they validate their findings, and why their methodology provides a reliable basis for identifying high-quality, actionable False Claims Act matters. The questionnaire converts those expectations into a checklist.

Key Details

Data miners who want a meeting with the Civil Fraud Section submit a white paper of no more than 10 pages total to [email protected], addressing seven items:

  • Scope. What type of fraud the analysis identifies, which federal programs, items, or services are the subject of it, and whether there is a geographic limitation.
  • Methodology. An overview of the analysis, including the types of data used, where the data came from, the assumptions incorporated into the model, whether subject-matter experts or clinicians were consulted, what testing was done on the model's accuracy, and what other external data supports the result. DOJ asks for specificity on sourcing, naming the database or dataset rather than the agency.
  • Legal hook. A summary of research into the statutes, regulations, or contracts that govern the conduct, and how the data analysis indicates violations of those requirements.
  • Alternative explanations. Whether there are potentially reasonable or legitimate explanations for the data signals, plus any analysis addressing them.
  • Best example. The single best instance of potential fraud the analysis identified, an explanation of why the signal indicates fraud as opposed to an administrative overpayment, and the estimated single damages to the federal government.
  • Filing history. A list of any qui tam cases previously filed, including case name with all defendants, and district.
  • The submitter. Background on the person or organization, relevant work experience and credentials, and any financial relationship with the industry that is the focus of the analysis. CVs may be attached.

DOJ says it will follow up to schedule a meeting if it concludes the analysis reflects a data signal with a reasonably strong correlation to potential fraud that is not already known to the government. That last clause does quiet work: on its face, a model that rediscovers a scheme the government has already identified does not clear the bar.

The department is careful to say such meetings are not a pre-filing requirement. Nothing in FOCUS blocks a relator from filing under seal without ever contacting the Civil Fraud Section. But DOJ says it will prioritize working with data miners that have demonstrated an investment in pre-filing diligence and commitment to analytical rigor, familiarity with program rules, and legally sufficient allegations, and the department retains intervention and dismissal authority. Sidley's False Claims Act Blog, which reported the release on July 20, 2026, wrote that the questionnaire puts concrete form to DOJ's expectations for data miner relators, who must show not merely an anomalous pattern, but a validated methodology, familiarity with the governing program requirements, consideration of alternative explanations, and evidence supporting fraud rather than error.

What It Means for Contractors

The seven questions are a defense outline written by the plaintiff's side of the house. Sidley's guidance to industry is to run them inward: government contractors, healthcare providers, and life sciences companies should evaluate their own data using these same standards, both to identify compliance risks and to anticipate how apparent outliers may be interpreted.

Three practical consequences follow. First, know where you are an outlier before someone else does. The metrics a data miner queries come from data the government already publishes or releases, which is exactly the material DOJ describes these relators as working from. Run the same queries against yourself and document why your numbers differ from peers.

Second, build the alternative-explanation record contemporaneously. Item four asks the relator to address innocent explanations, and item five forces a fraud-versus-overpayment distinction. Both are places contractors win cases, but only with evidence created at the time: the contracting officer's approval of a modification, the technical direction that changed a labor mix, the clinical documentation behind a coding pattern. A reconstruction assembled after a complaint unseals carries far less weight than a memo dated when the decision was made.

Third, treat the disclosure items as discovery targets. A relator who submitted a white paper has told DOJ its data sources, its assumptions, whether it consulted experts, how it tested accuracy, every prior case it filed, and any financial relationship with the industry it analyzed. Those admissions are attack surface for a defendant probing model validity, first-to-file and public-disclosure bars, and a relator's incentives. Contractors facing statistically driven complaints should ask early whether a FOCUS submission exists and press for it.

Sources