The 60-second answer
Bank statements are the heart of MCA underwriting. Funders look at 3–6 months of business bank statements and score them on revenue volume, revenue stability, NSF/overdraft incidence, average daily balance, deposit count and pattern, signals of other MCA debits (stacking), and revenue trend direction. The composite score drives both the approval decision and the factor rate offered.
In 2026, three things have changed materially: top funders now use automated bank-statement parsers (Ocrolus, Plaid, Heron Data) for near-real-time scoring; NSF tolerance has tightened by 1–2 incidents across all paper grades; and stacking detection has improved dramatically with same-day flagging of competing MCA debits.
What bank statement parsers actually do in 2026
The biggest single change in MCA underwriting over the past 24 months is the widespread adoption of automated bank-statement parsers. The dominant tools in 2026:
- Ocrolus: the market leader. Used by 40+ MCA funders for OCR-based parsing of uploaded statement PDFs. Extracts line-item transactions, classifies them (revenue, payroll, MCA debit, NSF fee, transfer), scores stability and trend.
- Plaid: bank-account-direct connection. Pulls real-time transaction data when merchant connects their bank account during application. Used by direct-app funders (OnDeck, Bluevine, etc.) and increasingly by ISOs with API-based submission.
- Heron Data: European entrant gaining US share. Specializes in revenue classification, recurring payment detection, and counter-party identification.
- In-house parsers: the top 10–15 funders have built internal bank-statement analysis tools that blend OCR, ML, and proprietary scoring. Often the most accurate but locked to their own deal flow.
What this means for merchants: bank-statement review is no longer done by a human reading 6 months of PDFs over coffee. It's an automated process that takes 5–15 minutes from upload to scored file. The output: a structured data record with revenue stats, NSF count, stacking signals, and a composite quality score.
What the parser actually measures
Revenue volume and consistency
Total deposit volume per month, classified as revenue (vs transfers, loan deposits, refunds). Most funders require minimum monthly revenue of $10,000–$25,000 depending on industry and advance size. The parser also measures consistency — coefficient of variation across months — with lower variation scoring better.
Average daily balance
The parser computes average daily balance across the statement window. Funders use this as a proxy for cushion. Higher average balance = better ability to absorb the daily ACH withdrawal post- funding. A merchant doing $40K/month revenue with $2K average balance is a much harder underwriting case than one doing $40K with $12K average balance.
NSF and overdraft incidence
Every NSF (non-sufficient funds) charge or overdraft incident is flagged. Modern parsers also detect "near-NSF" patterns where balance dipped close to zero but didn't trigger. Total NSF count across the statement window is one of the top 3 drivers of underwriting decision.
Typical 2026 tolerance bands:
- A paper: 0–2 NSFs in trailing 3 months
- B paper: up to 3 NSFs
- C paper: up to 4–5 NSFs
- D paper / rescue: up to 6–8 NSFs with reasonable explanation
- Hard decline: 9+ NSFs in 3 months, or any NSF in the most recent 7 days before application
These bands have tightened 1–2 incidents across the board vs 2024 norms. The cause: 2025 cohort data showed NSF count is a stronger default predictor than previously modeled.
Stacking signals
The parser scans for daily or weekly ACH debits that match the signature of other MCA payments. Known funder names get flagged directly (debits labeled "OnDeck," "Credibly," "CFG Merchant," etc.). Unknown funder names get scored against patterns — fixed- dollar debits hitting daily or weekly are flagged as likely MCA.
A clean file has zero MCA debit signals. A merchant with one open MCA shows up clearly. A merchant with 2+ open MCAs ("stacked") is usually an automatic decline at top-quartile funders and a defensive quote (much higher factor rate, smaller advance) at mid-tier funders.
The improvement in stacking detection is one of the most important underwriting shifts in 2026. The detection accuracy is now 90%+ at top funders, compared to maybe 60–70% in 2024. This is why stacked deals are getting declined more often — funders can see them now where they couldn't reliably before.
Revenue trend direction
The parser runs a simple regression across months in the statement window and produces a trend slope. Categories:
- Flat to slightly up (preferred): +0% to +15% month-over-month, low variance
- Modestly growing (also preferred): +5% to +20%, consistent direction
- Rapidly growing (raises questions): +25%+ — funder wonders if it's sustainable or driven by a one-time event
- Flat with high variance (acceptable for B/C paper): no trend but average revenue stable
- Declining (concerning): -5% to -15% — usually results in defensive pricing or decline
- Sharply declining (decline): -20%+ — almost always declined
Deposit frequency and pattern
The parser also looks at how deposits arrive. Daily or weekly card batch deposits (typical restaurant, retail) get one score. Weekly or bi-weekly contract payments (typical trucking, construction) get another. Monthly bulk deposits (typical professional services, medical) get another. The pattern matters because daily ACH repayment fits better against frequent deposit patterns.
How funders weight the composite score
Different funders weight the bank-statement composite differently, but a typical weighting model in 2026:
- NSF incidence: 25%
- Revenue volume: 20%
- Stacking signals: 20%
- Revenue trend: 15%
- Average daily balance: 10%
- Deposit pattern consistency: 10%
A file scoring 80+ on this composite typically gets A paper pricing regardless of other factors. A file scoring 60–80 is B paper. A file scoring 40–60 is C paper. Below 40 is usually D paper or decline.
Revenue volume thresholds in 2026
Minimum monthly revenue requirements have shifted modestly:
- A paper funders: $20,000+ monthly revenue minimum
- B paper funders: $15,000+ monthly revenue
- C paper funders: $10,000+ monthly revenue
- D paper / rescue funders: $7,500+ monthly revenue sometimes accepted with strong other factors
Note that these are minimums for approval, not optimum levels. To get the sharpest pricing at each tier, you generally want revenue 2–3x the minimum — A paper funders quoting sharpest factor rates usually want to see $40K+ monthly revenue, not $20K.
Industry-specific bank statement signals
Restaurants
Funders expect daily card batch deposits from Square, Toast, Clover, or processor of choice. Cash deposits should be roughly 15–35% of total revenue (industry-typical). Tips processed through payroll (vs cash) get scored more favorably. Funders often want to see at least 28 of 30 days with revenue deposits.
Trucking
Weekly or bi-weekly factoring deposits are normal. Direct broker ACH deposits show up at varying cadence. Funders look for stable monthly volume even if individual weeks are lumpy. Fuel card debits should be reasonable proportion of revenue (15–25%); excessive fuel spend relative to revenue raises questions about margin.
Construction
Lumpy deposits are expected — progress payments arrive irregularly. Funders run longer-window analysis (typically 6 months vs 3) to smooth out lumpiness. Large single deposits ($50K+) get verified against project documentation when possible.
Retail
Similar to restaurants but with more seasonal variation. Funders look at year-over-year comparison when possible to distinguish seasonal patterns from underlying trend.
Healthcare practices
Insurance reimbursements (Medicaid, Medicare, commercial) arrive in batches with varying delay. Funders look at deposit consistency quarterly rather than monthly. Patient direct-pay revenue is scored more favorably than insurance-driven revenue due to predictability.
How to position your bank statements before applying
Concrete actions that can shift your bank statement quality score:
- Pull statements 2–3 days before applying. Most recent activity matters most — don't submit statements that are 30+ days old.
- Label clearly with month and year on each PDF. Parsers handle this automatically, but human reviewers (still common for borderline files) appreciate clean labeling.
- Use one primary business account. Funders prefer to see consolidated revenue flow. If you have multiple accounts, disclose them and explain the structure.
- Avoid large transfers in the underwriting window. Moving $30K between accounts to boost average balance triggers parser flags. Looks like deposit smoothing, which is worse than having a lower honest balance.
- Don't let NSFs hit in the last 30 days. NSFs in the most recent window weight 2–3x more heavily than NSFs from 2 months ago. If you can avoid the last few NSFs by managing cash tighter for 30 days, do it before applying.
- Clean up labels on internal transfers. Many merchants label internal transfers with informal names that parsers misclassify as outflows or unknown debits. Use clean, unambiguous labels for self-transfers.
- Pay off small open MCAs before applying if possible. One small open MCA can flag stacking and decline the new application. Sometimes worth paying off (or paying down to under 50% balance) before applying for a larger amount.
What's tightening in 2026 vs loosening
Tightening
- NSF tolerance — tighter by 1–2 incidents
- Stacking detection — far more accurate, harder to slip a second open MCA past underwriting
- Recency weighting — last 30 days matters much more than 60+ days ago
- Revenue stability requirements — funders penalize high variance more than they did 18 months ago
Loosening
- Revenue minimum thresholds at C and D paper tiers — pushed down modestly to maintain origination volume
- Industry-specific accommodations — trucking, healthcare, and construction all have more nuanced underwriting frameworks than before
- Cash-heavy business treatment — funders now better understand cash deposit patterns at restaurants and retail
The takeaway
Bank statement quality is the single biggest factor in your MCA approval and pricing. 2026 underwriting is more automated, faster, and tighter on NSF and stacking than 2024 — but also more nuanced on industry-specific patterns. Knowing how funders will read your statements, and positioning the file accordingly, can shift your factor rate by 2–4 points. That's $1,000–$2,000 of fee on a typical $50K advance.
Pull your last 3 months of statements. Look at them with funder eyes. Count the NSFs. Check the trend. Find the stacking signals. Then decide whether to apply now, clean up first, or get matched to a funder whose criteria fit your specific statement profile.
Frequently asked questions
- What is bank statement quality in MCA underwriting?
- It's the composite score MCA funders assign to your last 3–6 months of business bank statements. It measures revenue volume, revenue stability, NSF/overdraft incidence, average balance, deposit consistency, stacking signals (other MCA debits), and trend direction. Almost every MCA approval/decline decision starts here.
- How has bank statement underwriting changed in 2026?
- Three major shifts: most top funders now use real-time bank-statement parsers (Ocrolus, Plaid, Heron Data) instead of manual review, NSF tolerance has tightened (most funders now decline at 4+ NSFs in 3 months, down from 6+ in 2024), and stacking detection has gotten dramatically better — same-day flagging of competing MCA debits is now standard.
- What's a typical NSF tolerance for MCA funders in 2026?
- A paper funders: 0–2 NSFs over 3 months. B paper: 3 NSFs. C paper: 4–5 NSFs. D paper / rescue: 6–8 NSFs with explanation. Above 8–10 NSFs almost always means decline regardless of revenue strength. This is materially tighter than 2024 norms.
- What revenue trend do MCA funders prefer?
- Stable to slightly growing. Funders run revenue regressions on 3–6 months of statements and prefer trends that are flat to up 10–15% over the period. Sharply rising trends raise questions (sustainable?); sharply declining trends often produce a decline. Volatile but flat-on-average trends are usually acceptable.
- How do I position my bank statements for the best factor rate?
- Pull recent statements 3 days before applying, label them clearly with month/year, run all business deposits through one primary account if possible, avoid moving large balances between accounts during the underwriting window, and don't carry NSFs in the latest 30 days. Clean statements often produce 2–4 points of factor rate improvement.