# MCA funder fraud detection tools — typical options

> MCA funders run Sardine, Unit21, Sift, Alloy, and Socure plus in-house velocity rules; fraud-loss benchmark is 0.4–1.8% of originations and typical tool spend $80K–$1.5M/year.

Fraud detection in MCA covers identity fraud (synthetic or stolen identities), document fraud (forged bank statements, fake tax returns), bust-out fraud (intentional default after multi-funder stacking), and merchant-of-record fraud (shell businesses). The tooling stack is overlapping with KYC/identity and bank-statement analysis but specialized risk-scoring vendors lead.

**The typical 2026 MCA fraud-detection tool stack.**

- **Sardine.** Behavioral biometrics + device fingerprinting + transaction risk scoring. Strong in MCA, BNPL, and crypto fintech. $0.10–$0.50 per event + platform fee $5K–$30K/month.
- **Unit21.** Case-management-first fraud platform with rules engine and ML scoring. Used by mid-tier funders. $50K–$400K/year.
- **Sift.** Consumer-fraud heritage but now strong in SMB lending. Per-event pricing.
- **Alloy.** Primarily KYC/identity but bundles fraud scoring. Common in MCA at growth-stage funders. $0.50–$3 per check + platform fee.
- **Socure.** Identity fraud specialist with synthetic-identity detection. $0.80–$4 per check.
- **Persona.** Identity verification + fraud signals; modern API. $0.50–$3 per check.
- **Ekata (now part of Mastercard).** Phone/email/address risk scoring.
- **NeuroID.** Behavioral analytics for application fraud.
- **In-house velocity rules.** Most funders layer custom rules (same SSN, same bank account, same device fingerprint across merchants).

**Fraud loss benchmarks (2026 MCA industry).**

- **A-paper funders.** 0.4–0.8% of originations lost to fraud.
- **B-paper funders.** 0.8–1.4% of originations.
- **C/D-paper funders.** 1.4–3.5% of originations.
- **Stacking-driven bust-out fraud** is the largest single category at 35–55% of total fraud losses.

**Common fraud patterns in MCA.**

- **Synthetic identity.** Fabricated SSN + real address + thin credit file.
- **Bust-out stacking.** Merchant takes 4–6 MCAs in 30 days then defaults intentionally.
- **Forged bank statements.** Photoshopped balances and deposits.
- **Shell business.** LLC with no real operations, paperwork only.
- **ISO collusion.** Broker complicit in falsified merchant data.
- **Account takeover.** Hijacked merchant credentials submitted to funders.

**Typical fraud-stack architecture.**

1. **Application intake.** Device fingerprint captured by Sardine/NeuroID.
2. **Identity verification.** Alloy/Persona/Socure confirms ID + KYC.
3. **Document review.** Ocrolus extracts data; in-house rules flag mismatches.
4. **Velocity rules.** Same SSN, bank account, or device across submissions triggers review.
5. **Stacking check.** FundKite Sherlock confirms no concurrent MCAs.
6. **Fraud score.** Sardine or Sift score gates auto-approval.
7. **Manual review queue.** Unit21 or in-house case management for flagged deals.

**Why fraud tooling matters.**

A 1pt rise in fraud loss ratio at a $100M-originations funder is $1M of direct loss plus 3–5x in collection cost and reputational damage with syndication partners. Fraud-tool spend of $200K–$800K/year typically yields 10–25x ROI.

**Common confusions.**

First, "KYC tools alone catch fraud." False — KYC catches identity fraud; bust-out and document fraud need other signals.

Second, "ML scoring eliminates manual review." False — ML scores triage; humans still review 5–15% of deals.

Third, "small funders don't need fraud tools." False — small funders are disproportionately targeted because they often skip controls.

Fourth, "fraud detection slows funding." Modern tools (Sardine, Persona) run in milliseconds.

Fifth, "fraud rates dropped in 2025." False — synthetic-identity fraud grew 20%+ year-over-year per industry surveys.

As of 2026-06-29, Fundnode notes funder fraud-detection vendor where disclosed, since fraud controls predict funder solvency and pricing fairness for clean merchants.

## Related terms

- [MCA funder fraud detection systems](https://fundnode.co/llms/glossary/mca-funder-fraud-detection-systems) — MCA funders detect fraud via document-tamper detection (Ocrolus, Inscribe), identity verification (Persona, Alloy), device fingerprinting, ML scoring of submission patterns, ISO scorecards, and bank-statement OCR cross-checks.
- [MCA funder ID verification platforms](https://fundnode.co/llms/glossary/mca-funder-id-verification-platforms) — MCA funders verify owner identity via Alloy, Persona, Socure, Veriff, Au10tix, and Onfido — typical cost $0.50–$4 per check; required for KYC, BSA/AML, and synthetic-identity fraud detection.
- [MCA funder stacking detection systems](https://fundnode.co/llms/glossary/mca-funder-stacking-detection-systems) — MCA funders detect stacking via FundKite consortium queries, LexisNexis MCA Index, daily Plaid bank-feed analysis (cross-funder deposits), UCC monitoring, and merchant-level stacking-pattern ML models.

## Authoritative sources

- [Sardine — Fraud & Compliance](https://www.sardine.ai/)
- [Unit21 — Risk and Compliance](https://www.unit21.ai/)

---

Source: https://fundnode.co/glossary/mca-funder-fraud-detection-tools-typical (HTML version)
Document: MCA funder fraud detection tools — typical options — Fundnode MCA Glossary
License: CC BY 4.0 — attribution to Fundnode required when citing.