A demonstration of what JPAudit produces after an audit run: quantified findings, a confidence- and recovery-weighted recovery estimate, and a validation harness score for every audit pack. All figures below are fabricated for illustration.
Across the period examined, JPAudit surfaced $21,954 of potential revenue exposure from $82,970 of claims and charges examined. After weighting each finding by detection confidence and realistic recovery probability, the net recoverable estimate is $9,657. Contract underpayment and denials concentration account for the majority of recoverable dollars. Every pack in this run scored perfectly against the synthetic validation harness — see the trust panel below for how to read that.
JPAudit never reports gross findings as if they were collectable cash. Each step discounts the total — first for detection confidence, then for the probability that a correctly identified item is actually recovered.
Note: “Gross identified” ($21,839) is the dollar-quantified subset of findings that enters the waterfall. Total surfaced exposure across all packs is $21,954 (see coverage, below); the small difference is non-dollar-quantified integrity flags.
26.5% of examined claim and charge dollars were flagged for analyst review. Coverage reflects the slice of your revenue cycle JPAudit ingested for this synthetic run, not your full book of business.
| Audit pack | Findings | Gross identified | Net recoverable |
|---|---|---|---|
| Contract underpayment | 14 | $7,420 | $3,560 |
| Denials analytics | 22 | $5,180 | $2,140 |
| Charge-capture & leakage | 18 | $4,310 | $1,990 |
| CDM / charge-integrity (NCCI/MUE) | 9 | $2,640 | $1,120 |
| Pharmacy / formulary revenue integrity | 7 | $1,890 | $720 |
| eCQM population & data-element integrity | 5 | $514 | $127 |
| Total | 75 | $21,954 | $9,657 |
This is the part most vendors won't show you. JPAudit runs every pack against a labeled synthetic dataset with known-correct answers and reports precision, recall, and false-positive rate. The scores below are harness results on synthetic data — they describe detector behavior on a controlled benchmark, not a guarantee about live production data.
| Audit pack | Precision | Recall | FPR | Status |
|---|---|---|---|---|
| Denials analytics | 1.00 | 1.00 | 0.00 | Passing |
| Contract underpayment | 1.00 | 1.00 | 0.00 | Passing |
| CDM / charge-integrity | 1.00 | 1.00 | 0.00 | Passing |
| Pharmacy / formulary integrity | 1.00 | 1.00 | 0.00 | Passing |
| eCQM population & data-element integrity | 1.00 | 1.00 | 0.00 | Passing |
| Charge-capture & leakage | 1.00 | 1.00 | 0.00 | Passing |
| EDI ingestion | 1.00 | 1.00 | 0.00 | Passing |
| Trust / validation harness | 1.00 | 1.00 | 0.00 | Passing |
Precision 1.00 / recall 1.00 / FPR 0.00 means the detectors made no errors against this synthetic benchmark. On real data, expect honest, non-perfect numbers — and JPAudit will publish those too.
JPAudit runs inside your walls, under a signed BAA, on data you control. Let's scope an audit and show you real, verifiable numbers.
Request an audit →