MCA funder volatility pricing model explained.

The 60-second answer

Modern MCA underwriting — driven by parser-stack tools like Ocrolus, Plaid, and Heron Data — scores every applicant on revenue volatility, not just average revenue. The volatility model uses five inputs:

• Coefficient of variation (CV) on monthly deposits
• NSF and overdraft frequency and pattern
• Trend slope (regression on trailing 12 months)
• Day-of-week and intra-month deposit variance
• Seasonality fit against the funder's industry benchmark

Two merchants at the same average monthly revenue can price 6–10 basis points apart based purely on the volatility score. Understanding the model lets you time the application — and clean up the volatility inputs — before the funder pulls statements.

Coefficient of variation (CV) — the headline metric

CV is the most-weighted volatility input in 2026 underwriting. The math is simple:

CV = (standard deviation of monthly deposits) / (mean of monthly deposits)

Expressed as a percentage. Two worked examples:

• Low-CV merchant: 6 months of deposits at $30K, $32K, $28K, $31K, $29K, $30K. Mean $30K, std dev $1.4K. CV = 4.7%. Tier-A volatility score.
• High-CV merchant: 6 months at $30K, $45K, $15K, $40K, $20K, $30K. Mean $30K, std dev $11.7K. CV = 39%. Tier-C volatility score.

The 2026 CV pricing bands at most major funders:

• CV under 15%: Tier A volatility; factor improvement of 2–3 basis points
• CV 15–30%: Tier B volatility; neutral
• CV 30–50%: Tier C volatility; factor penalty of 3–5 basis points
• CV over 50%: Tier D volatility; factor penalty of 6–10 basis points or decline

NSF cycles and overdraft patterns

NSF and overdraft events are the single highest-correlated predictor of MCA default in funder portfolio data. Funders score on three NSF dimensions:

• Frequency. Total NSF count in trailing 90 days. Zero is the only "clean" score; 1 NSF drops you a partial tier; 3+ closes the funder door at most major funders.
• Pattern. Recurring NSFs (1st-of-month rent, 15th-of-month payroll) score worse than scattered NSFs because they signal structural cash-flow gap.
• Recency. An NSF in the last 30 days is heavily weighted; NSFs older than 60 days fade in the model.

Overdraft fees (paid items that exceeded balance) score similarly but with somewhat less weight than NSFs (returned items). Most parser-stack tools tag both events in the same volatility score.

Trend slope — the regression line

Underwriters fit a least-squares regression line to trailing 12 months of monthly deposits and read the slope (the trend). The slope dimension:

• Strong positive slope (10%+ month-over-month growth): Often offsets a full tier of CV penalty. The model interprets high CV with positive slope as growth volatility, not distress.
• Flat slope (-3% to +3% per month): Neutral.
• Mild negative slope (-3% to -8% per month): One-tier penalty on top of the CV score.
• Steep negative slope (-8%+ per month): Two-tier penalty; some funders decline regardless of average revenue.

The 12-month window is critical. A merchant who had a great year then a soft 3 months will look different in the 12-month regression than in the 6-month average. Time your application so the trend slope reflects the current trajectory you want the underwriter to see.

Day-of-week and intra-month variance

Beyond month-to-month variance, parser-stack tools now look at intra-month deposit patterns. Three observations matter:

• Day-of-week concentration. Restaurants with 60% of weekly revenue on Friday + Saturday have higher day-of-week variance than a service business with even Monday-Friday revenue. The day-of-week variance feeds into the daily-ACH risk model because daily ACH on a slow Tuesday is harder to absorb than on a busy Friday.
• Pay-period clustering. Service businesses billed monthly (consulting, agencies) often have 60%+ of monthly revenue on a single deposit day. This pattern scores worse than continuous revenue because daily ACH is brutal on the 25 zero-deposit days.
• Card-vs-ACH split. Restaurants and retail with high card share are scored against the more predictable daily-batch deposit pattern; B2B businesses with check-and-wire revenue are scored differently because deposit timing is lumpier.

Seasonality fit

Industry-specialty funders (CFG, Credibly, Reliant) maintain seasonality benchmarks for each industry. The underwriter compares your deposit pattern against the industry benchmark and scores the fit:

• Strong fit (within 10% of industry benchmark). The model interprets this as normal industry seasonality, not idiosyncratic distress. Pricing neutral.
• Misfit (deviation from industry benchmark). The model interprets this as idiosyncratic risk and adds a penalty. A landscaper with high December revenue is a misfit and gets a penalty; a landscaper with low December revenue fits and gets neutral.

The seasonality model is why industry specialty funders price tighter than generalist funders — they have better benchmarks and can distinguish normal industry variance from true risk.

The parser stack that drives the 2026 model

The volatility model has matured because funders adopted bank-statement parser tools. The 2026 parser stack:

• Ocrolus. Dominant in mid-market and sub-prime MCA underwriting; extracts deposit and NSF data with high accuracy across most bank statement formats.
• Plaid. Direct bank data feed (not PDF parsing); gives real-time balance and transaction data. Used by marketplace lenders (Bluevine, Fundbox, OnDeck).
• Heron Data. Specialty in transaction categorization; identifies revenue vs transfers vs other deposit types.
• MX (Money Experience). Bank-direct API similar to Plaid; used by some marketplace lenders.

The parser stack matters to merchants because mis-classified deposits can hurt your score. Cash deposits labeled "ATM Deposit" sometimes score as transfers (not revenue); rent payments to a separate account that are then transferred back can show up as internal transfers that suppress your apparent revenue. Always submit statements with clean deposit labeling.

Worked example: same revenue, different volatility scores

Two merchants at identical $40K/month average revenue, identical FICO, identical industry, identical time-in-business. Different volatility profiles produce dramatically different pricing:

Merchant A — low volatility

• Monthly deposits: $38K, $41K, $40K, $42K, $39K, $40K
• CV: 3.6% (Tier A)
• NSFs in 90 days: 0
• Trend slope: +1% per month (flat)
• Industry seasonality fit: within 8% of benchmark
• Underwriter pricing: 1.24 factor on 12-month term

Merchant B — high volatility

• Monthly deposits: $25K, $55K, $30K, $50K, $35K, $45K
• CV: 30% (Tier C)
• NSFs in 90 days: 2
• Trend slope: 0% per month
• Industry seasonality fit: 15% deviation from benchmark
• Underwriter pricing: 1.34 factor on 8-month term

Same average revenue, 10 basis points apart on factor, and a shorter term to boot. On a $40K advance, that's $4,000 in additional fees — and a daily ACH that's 40% higher because of the shorter term.

Three concrete moves that improve volatility score

1. Consolidate banking into a single primary account

Multiple operating accounts produce transfer noise that the parser stack often misclassifies. Money moving between your own accounts can show up as transfers (good) or as deposits-and-withdrawals (bad — inflates volatility). Move to a single primary operating account 90+ days before applying.

2. Build a 90-day clean banking record

Zero NSFs, zero overdrafts, daily positive balance, average daily balance trending up. This single change moves more merchants up a tier than any other lever. The parser tools pull trailing 90 days as the primary window, so 90 days of clean banking can fully erase a single NSF from the prior period.

3. Time the application to a stable trailing window

If you had an anomalous month (a closure week, a big one-time payment, a temporary revenue dip), wait for it to age out of the trailing 6-month window. Submitting in your worst possible month skews the CV calculation. Submitting after 3 stable months erases the anomaly.

What sophisticated merchants ask the broker before submitting

Three questions to ask before the broker submits your file:

• What parser tool does the funder use? Knowing whether it's Ocrolus, Plaid, or Heron tells you what the underwriter actually sees and lets you submit optimally.
• What's the funder's published CV pricing band? Some funders publish their volatility model; most don't. If the broker doesn't know, ask them to ask the funder rep before pulling the file.
• How does the funder handle seasonality fit? Industry-specialty funders use industry benchmarks; generalist funders penalize all variance equally. Routing to the right funder type matters.

The bottom line

The 2026 MCA underwriting model is much more sophisticated than the 2020 average-and-FICO score. Volatility matters as much as revenue. Two merchants at the same average can price 10 basis points apart based purely on CV, NSF, trend, and seasonality fit. Understanding the model — and cleaning up the inputs for 90 days before applying — is the highest-ROI move most merchants can make. It's free, takes only operational discipline, and can save 4–8 basis points on the next funding round.