Case studies
VeraStream 1 worked · production rollouts

VeraStream first worked audit case: from a 25-to-60 quarterly sample to 95% continuous coverage.

Real production rollouts, real before/after coverage shifts, and the workpaper shape your external auditor signs off against. 95% of every transaction, every business day, before the wire clears — replacing the legacy 25-to-60-line quarterly sample that legacy review never had the runway to find.

Frequently asked

/case-studies FAQ — population vs sample, the workpaper shape, and what ships

The five questions the case studies surface answers most: what a case covers, which detectors lead the holds, whether the data is real, how to run the same on your ledger, and the relationship between the 95% benchmark on /case-studies and the announcement blog post. Plain HTML answers — no JavaScript required to read.

What does a VeraStream case study cover?

Each worked example walks the before/after on a real ledger — the legacy sample-based review the buyer is replacing, the production detector rollout that replaced it, and the workpaper shape the external auditor signed off against. The 95%-miss benchmark, the 25-to-60-line quarterly sample, and the 47-day lag from disbursement to reviewer desk are the population-vs-sample numbers every case study anchors on.

Which of the eight production detectors lead holds most often?

detectDuplicatePayment and detectVendorRisk lead the majority of holds in the first 90 days of any rollout — a duplicate-invoice ring tripping findDuplicateInvoices and a vendor typosquat ring tripping detectVendorRisk together account for the largest single-day amount recovered. The full production set — evaluatePolicy, findDuplicateInvoices, detectExpenseAnomalies, detectVendorRisk, detectThresholdGaming, detectRoundDollar, detectDuplicatePayment, detectGhostEmployee — runs continuously, with each flag lifting into the same workpaper envelope.

Are the case studies real production data or anonymized?

The numbers, the before/after coverage shifts, and the workpaper shape are pulled from production-tenant rollouts. Vendor names, amounts, and approvers are anonymized — the case study surfaces the population-level signal (95% coverage, 25-to-60 sample size, 47-day lag) rather than individual-finding PII. The same eight detectors run on your ledger unchanged at /audit.

How do I run the same eight detectors on my ledger?

Drop a CSV of AP rows at /audit and the same eight detection functions run in your browser — file stays in your tab, no upload, no signup. For continuous coverage against your live ERP, browse the pricing tiers — Starter, Continuous, or Enterprise — that launch those same detectors against NetSuite, SAP, Coupa, Workday, or any CSV-shaped export.

Is the 95% benchmark on /case-studies the same as the announcement blog post?

Yes. The 95%-vs-5%-coverage framing on the announcement blog post (/blog/announcing-seven-detector-verastream-rollout) and the case studies index is the same single shift — VeraStream evaluates 95% of every transaction, every business day, before the wire clears, while the remaining 5% is the legacy sample-based review the rollout replaces.

Run the same on your ledger

Drop a CSV at /audit — or launch the eight continuously at /pricing.

The same eight production detectors that produced each case study run on your AP ledger at /audit — no signup, no sales call required to see first findings. Pick Continuous or Enterprise at /pricing to launch them continuously against your live ERP. For the workpaper-integrity posture the case-study workload rests on, see /security.