# Medical Bill Review Software > Medical bill review software: every billed line compared to UCR and geographic benchmarks, outliers flagged with percentile context. Humans decide. Canonical page: https://medrecords.ai/product/medical-bill-review/ --- NewWe now read the actual MRI and CT imaging — not just the radiologist's 1-page report. [See how →](https://medrecords.ai/product/imaging/) [Home](https://medrecords.ai/) ›[Product](https://medrecords.ai/product/) › — Billed-Charge Benchmarking MEDICAL BILL REVIEW SOFTWARE ## The source exists. Is the amount reasonable? Medical bill review software compares every billed line to UCR and geographic benchmark data, flagging outliers with percentile context rather than a bare "too high." The ledger proves a charge has a source; this proves whether the number belongs. Flags inform negotiation — they never auto-reduce, reprice, or determine payment. The pricing decision stays human. Flags inform negotiation — they never auto-reduce or reprice a bill. The pricing decision stays human. [Test a file ](https://medrecords.ai/test-a-file/?src=product-medical-bill-review) [Book a demo](https://medrecords.ai/demo/) Adams, T. · Case #IME-4812 Benchmark review · 2 packets Charge — Billed — Vs. benchmark Ortho consult — 4/02 $ ▮▮▮ IN RANGE Physical therapy — visit 3 of 7 $ ▮▮ IN RANGE Imaging read — right knee $ ▮▮▮ ABOVE RANGE Each flag — **source page + benchmark cited** One outlier flagged Every line Compared to a named benchmark Percentile context On every flag, never a bare label Dual citation Source page plus benchmark ### Benchmarked against the right market, not a national average. UCR benchmark bill review only means something if the benchmark fits: the same procedure carries different typical pricing in different markets. Every billed line is matched to usual, customary, and reasonable data for its procedure and its billing geography, and the comparison shows which benchmark it was drawn from. UCR and fee-schedule references, matched per line Geography-matched, and the locale is shown One charge · its comparison basis Billed line on the provider statement — p.14 Procedure matched to the benchmark set — UCR Locale of the billing provider applied — GEO-MATCHED The basis is named on every line — **never a black box** . Where the charge sits — vs. benchmark range The flag shows **how far outside the range** a charge sits — not just that it does. ### Percentile context, not a bare "too high." A billed charge reasonableness analysis that just says "high" doesn't survive the first phone call. Every flag here shows where the charge falls against the benchmark distribution for that procedure in that area, so your reviewer can say how far outside the typical range it sits — the part that actually carries weight in a negotiation. Outliers ranked by distance from the typical range In-range lines confirmed too — silence isn't a verdict ### 2 citations on every flag. Medical bill audit AI is only as good as its receipts, so each flag carries a dual citation: the source page in the file where the charge appears, and the benchmark it was compared against. One click verifies the charge is real; the second shows the basis for calling it an outlier. It pairs naturally with the [Claims Billing Ledger](https://medrecords.ai/product/claims-billing-ledger/) — the ledger proves the source, this weighs the amount. Source-page citation on the charge itself Benchmark citation on the comparison One flag · both receipts Imaging read — flagged above range — OUTLIER Charge as billed, in the file — p.140 Benchmark set and locale used — UCR · GEO Both sides of the comparison stay **inspectable** . What the flag is · what it is not Reasonableness signal, percentile context, dual citation DELIVERED Auto-reduced line, repriced bill NEVER Payment determination, coverage decision NEVER AI flags. Your reviewer decides. The boundary ### A signal for the negotiation — never a repricing. The output of this review is evidence: a cited, percentile-ranked picture of which charges sit outside their market. What gets paid, disputed, or negotiated is your call, made by your adjuster, attorney, or bill reviewer. Audit-grade and source-linked on every flag — that is what makes the position defensible when it's challenged. [See Claims Billing Ledger ](https://medrecords.ai/product/claims-billing-ledger/) ### From billed lines to a benchmarked review. 3 steps between the packet and outliers your reviewer can act on. 01 Upload the file Bills, statements, and EOBs — mixed in with the treatment records is fine. 02 Every line benchmarked Each charge matched to UCR data for its procedure and geography; outliers flagged with percentile context. 03 Work the flags Your reviewer clicks through the dual citations and decides what each outlier means for the negotiation. ### Who benchmarks bills with it. Anyone who has to answer the same question: the charge is real, but is the number right? [ Insurance carriers Demand specials weighed against market data before the reserve is set or the offer goes out. For carriers ](https://medrecords.ai/solutions/insurance-carriers/) [ TPAs & case management A consistent reasonableness screen across clients, with the basis documented on every flag. For TPAs ](https://medrecords.ai/solutions/tpas/) [ Law firms Know which of your specials will draw fire, and which of theirs deserve it — before the table. For law firms ](https://medrecords.ai/solutions/law-firms/) FAQ ### Bill review, answered. Billed charges are compared against usual, customary, and reasonable (UCR) benchmark datasets and fee-schedule references, matched to the procedure and the billing geography. Every flag names the benchmark it was compared to, so the basis is inspectable rather than a black box. Benchmarks are matched to the geographic area where the service was billed, not a flat national average. The same procedure has different typical pricing in different markets, and the comparison respects that: each line shows the locale its benchmark was drawn from. No. The output is a reasonableness signal with percentile context and citations — a flag for your negotiator or bill reviewer to weigh. It never reprices a line, recommends a payment amount, or issues a payment determination. The pricing decision stays human. Every flagged charge shows where it falls against the benchmark distribution for that procedure in that area, instead of a bare high or low label. Your reviewer sees how far outside the typical range a charge sits, which is what actually carries weight in a negotiation. Each flag carries 2 citations: the source page in the file where the charge appears, and the benchmark it was compared against. One click verifies the charge is real; the second shows the basis for calling it an outlier. Both sides of the comparison stay inspectable. ### Related capabilities. The billing side of the same cited file — source-proof, roll-ups, and demand checks. [ Claims Billing Ledger Every billed amount traced to its source page — including the ones without one. Explore ](https://medrecords.ai/product/claims-billing-ledger/) [ Medical Billing Summary A 1-page rolled-up billing summary grouped by provider, date, or category. Explore ](https://medrecords.ai/product/medical-billing-summary/) [ Demand Package Audit Every claim and billed figure in a demand, checked against the record. Explore ](https://medrecords.ai/product/demand-package-audit/) [ Fraud & Inconsistency Signals Inconsistencies across the file surfaced with citations — signals, not verdicts. Explore ](https://medrecords.ai/product/claims-fraud-signal-detection/) ### Benchmark one of your own files. Upload a file and get back a sample review — every line benchmarked, outliers flagged with percentile context. Handled under our BAA; never used to train a model. [Test a file ](https://medrecords.ai/test-a-file/?src=product-medical-bill-review) [Book a demo](https://medrecords.ai/demo/) [All capabilities ](https://medrecords.ai/product/)