Why Two Businesses With the Same Numbers Get Different Offers

Two operators submit otherwise identical files. One runs a 20-year-old auto repair shop. The other runs a two-year-old vape retailer. The auto repair shop gets a clean approval at standard pricing. The vape retailer gets either declined outright or quoted noticeably higher — on the same revenue and the same bank statement quality.

The difference isn't bias. It's industry risk scoring, and it sits inside almost every working-capital underwriting model. Most operators experience it as opaque or unfair; in reality, it's driven by three measurable drivers that the desks track over multi-year portfolio performance.

Here's what those three drivers actually are, and what they mean for which industries get easier capital.

Driver 1: portfolio default rate by industry

The first and largest input is observed default behavior in the desk's own portfolio. Every working-capital lender tracks repayment performance segmented by industry classification (typically NAICS code or a custom industry mapping).

Industries with consistently elevated default rates over multi-year windows get priced higher and sized smaller. This isn't moral judgment; it's portfolio math. If an industry runs at a 12% default rate when the rest of the portfolio runs at 4%, the deals from that industry have to be priced to compensate the additional loss expectancy. There's no other way for a portfolio to survive over time.

What sits in the elevated-risk bucket varies by lender, but commonly includes: certain hospitality categories (especially nightlife), some construction subspecialties (particularly residential roofing in storm-event regions), trucking with limited time-in-business, and recently emerged categories without sufficient portfolio history (cannabis, certain crypto-adjacent businesses).

What sits in the lower-risk bucket: established service businesses (auto repair, professional services, healthcare), mature retail with 10+ years of history, and established manufacturing.

This driver accounts for roughly 60% of industry risk pricing variation. It's the largest input and the hardest for an operator to influence directly.

Driver 2: operational fragility under stress

The second input is whether the industry's revenue model continues to function during predictable disruption events.

Some industries have revenue models that bend without breaking — a dental practice can defer non-emergency procedures during a slow week and most of the revenue still happens, just shifted in time. A subscription-revenue SaaS business has revenue largely uncoupled from any one bad operational week.

Other industries have revenue models that snap under stress. A wedding venue has very specific, calendar-bound revenue events that cannot be shifted. A weather-dependent outdoor service has revenue that disappears entirely during weather events. A category dependent on specific large clients (a 3-year contract that's the majority of revenue) has revenue that can vanish at contract expiration.

Underwriters score this as a stress-resilience signal. Industries whose revenue can absorb a 4-week disruption get better pricing. Industries whose revenue model is fragile to predictable disruptions get priced to reflect that fragility.

This driver accounts for roughly 25% of industry pricing variation. It's particularly important for newer businesses without long operating history, where the industry signal is doing more work because the business-specific signal is thinner.

Driver 3: regulatory and reputational risk to the lender

The third input is straightforward but rarely discussed openly. Some industries create regulatory exposure or reputational concern for the lender itself, regardless of the specific applicant's behavior.

Industries that touch federally illegal activity (cannabis at the federal level, certain crypto activities), industries with high consumer-complaint risk profiles (some debt collection, some short-term consumer lending), and industries with elevated AML or sanctions concerns (certain money-services businesses, some international remittance) get priced higher or excluded entirely because of the cost of compliance and the headline risk to the funder.

This driver accounts for roughly 15% of industry pricing variation. It's also the one most likely to result in a flat decline rather than a higher-priced offer, because the calculus is binary at most desks: either the regulatory exposure is acceptable or it isn't.

What an operator can actually do

Industry classification is largely fixed by what your business does. But the file-level signals that adjust pricing within an industry are real.

1. Document long history within a high-risk industry. A 12-year-old hospitality business gets meaningfully better terms than a 2-year-old one in the same category. The operator who has survived 12 years has demonstrated something the industry-level data can't capture about their specific operation.

2. Show diversified revenue within the industry. A construction subcontractor with one GC representing 80% of revenue is fragile. The same business with seven active GCs and no single GC above 25% of revenue is structurally different. The bank statements will show this; underwriters read it.

3. Be explicit about the off-cycle plan. If your industry has known seasonal weakness (landscaping in winter, retail in February), volunteer the explicit plan for those months in a one-paragraph note attached to your application. "Q1 typically runs at 60% of Q3 revenue; we maintain a 12-week cash buffer entering January and run reduced operating expense through the trough" is a sentence that moves pricing.

4. Provide industry-specific context for any anomaly. A bank statement event that looks unusual to an algorithmic model is often perfectly normal in your specific industry. A roofing contractor with three weeks of zero deposits in February isn't a problem; it's a known seasonal floor. Saying so in writing prevents the file from being scored against the wrong baseline.

5. Know which lenders are friendlier to your industry. Different desks have different industry concentrations and different risk appetites. A lender whose existing portfolio is 18% restaurant has different appetite for restaurant deals than one whose portfolio is 3% restaurant — in both directions, depending on whether they're trying to grow or trim that exposure. There's no public list of who's where on this; the practical workaround is applying to two or three reputable shops and reading the response patterns.

What you can't do

You can't materially change your industry classification. NAICS code shopping ("we'll just say we're 'business services' instead of 'used car dealer'") is a fast way to invalidate a funded deal. The truthful answer to "what does your business do" governs the underwriting model. Any other answer creates a contractual problem that's worse than the pricing problem you were trying to solve.

You also can't argue your way out of a category-level decline. If a lender doesn't fund your industry, the conversation is closed at intake. Find a lender that does.

The bottom line

Industry risk is real, measurable, and reflected in pricing. It isn't punitive; it's portfolio math derived from observed performance over years of data. Operators in higher-risk industries who demonstrate the operator-level signals above routinely get priced inside their industry's range — sometimes meaningfully better than the industry mean.

The shortest version: the industry signal is the starting point, not the ending point. The file-level signals you control move you within that range. Operators who treat industry risk as a verdict rather than a baseline tend to under-shop their offers and accept worse terms than they could have negotiated. Don't.