Every week, a referral partner sends over a deal they’re confident in. Good business. Real revenue. Owner has skin in the game. And somewhere between the handshake and the approval, it dies.
Not because the business wasn’t fundable. Because the file wasn’t readable.
The SBA has rules. A lot of them. But the real reason most files miss isn’t buried in the SFO or the SOPs. It’s simpler than that: lenders are pattern-matching machines, and most files don’t give them the pattern they’re looking for.
Let me show you what they’re actually looking for, and where the gaps usually are.
What the SBA Is (And Isn’t)
First, a clarification that matters. The SBA doesn’t lend money. It guarantees loans made by approved lenders. That distinction changes everything about how you approach a submission.
When you submit a file, you’re not convincing the SBA. You’re convincing the lender who has to put their capital at risk first, knowing the SBA guarantee kicks in only if things go sideways, and only under specific conditions. That lender has their own credit culture, their own risk appetite, and their own interpretation of what the SBA will and won’t back.
Which means there’s no single SBA underwriter somewhere reading your file. There are dozens of lenders, each running their own shop, each with their own preferences layered on top of the SBA’s baseline requirements.
What the SBA does establish is the floor. And most files don’t clear it, not because the business is bad, but because the narrative is incomplete or contradictory.
The Four Things Every SBA Lender Is Actually Evaluating
Forget the checklist for a second. When a credit officer opens your file, they’re answering four questions. Everything else is supporting evidence for one of these four.
1. Can This Business Repay the Debt?
This is the primary question. The SBA’s general standard is 1.25x DSCR on a global basis. That means the business’s net operating income, after adding back depreciation and amortization, needs to cover all debt payments, including the proposed loan, by at least 125%.
Where files break down: owners who take distributions that gut the income picture, or businesses where the tax return tells a completely different story than the bank statements. Lenders reconcile these. If your file can’t explain the gap, you’ve already lost the argument.
Also: global cash flow includes the owner’s personal debt load. If your borrower is carrying $4,000 a month in personal obligations and the business only generates enough to cover its own debt at 1.3x, a lender running the global DSCR may come up short. You need to know this before you submit, not after the decline.
2. Is There Collateral to Fall Back On?
The SBA requires lenders to take all available collateral when it’s available. That includes business assets, real estate, even personal assets in some cases. But here’s what most people get wrong: collateral is not the deciding factor. It’s a risk mitigant, not a primary qualifier.
A deal won’t get approved because there’s great collateral. But it can get conditioned or declined if available collateral isn’t properly documented and offered. Don’t let a clean deal stumble because the asset schedule was vague or the real estate wasn’t properly valued.
3. Who Is This Borrower?
Credit score matters, but it’s not the whole picture. Character matters, and the SBA is explicit about this. What a lender is looking for: does the borrower have a history of meeting obligations? Have they disclosed everything? Is there criminal history that triggers ineligibility? Do they have relevant industry experience?
The file needs to tell a coherent story about who this person is. Resume, personal financial statement, background summary — these aren’t bureaucratic exercises. They’re your chance to answer the character question before the underwriter starts asking it.
4. Does the Loan Purpose Make Sense?
Use of proceeds needs to be SBA-eligible and clearly documented. “Working capital” is not a use of proceeds. “Replenish operating cash depleted during Q3 due to a 90-day contract gap with primary client” is a use of proceeds.
Lenders want to see that the capital is going somewhere that makes the business stronger or more stable, not somewhere that plugs a hole that’s going to reopen. If the use of proceeds is murky, the deal is murky.
Where Most Files Actually Break Down
I’ve worked through enough of these to know the patterns. Here’s where the friction consistently lives:
| The Problem | What Lenders See | What Fixes It |
|---|---|---|
| Tax returns show a loss; owner says business is profitable | Contradiction without explanation | CPA letter explaining add-backs with documentation |
| Multiple businesses with shared expenses | Unclear who’s actually carrying the debt | Global cash flow analysis with allocation memo |
| Owner salary far below or above market | Income manipulation, real or perceived | Compensation justification tied to market comps |
| Recent spike in revenue (last 12 months) | Unsustained growth trend | Client contracts, pipeline, and explanation of drivers |
| Significant personal debt on PFS | Global DSCR at risk | Run the number before submitting; restructure if needed |
| Vague use of proceeds | No clear repayment story | Specific, documented narrative tied to business need |
| Missing or incomplete entity documents | Can’t confirm ownership structure | Full organizational documents, operating agreement, ownership schedule |
Notice that most of these aren’t financial problems. They’re documentation and narrative problems. The money works. The story doesn’t.
The Eligibility Layer That Trips Up Good Deals
There’s a whole category of declines that have nothing to do with credit quality. The business simply isn’t eligible, or the deal structure isn’t eligible, or someone in the ownership chain creates an issue.
Some of this is obvious. Real estate holding companies, passive businesses, financial businesses, nonprofits – the SBA has a list of ineligible business types. Most experienced referral partners know this.
What trips people up is the less obvious eligibility layer:
- The 20% ownership rule. Anyone owning 20% or more must personally guarantee the loan. If there’s a complex ownership structure with multiple tiers of entities, tracing who actually owns what, and who therefore needs to guarantee, can get complicated fast. If this isn’t sorted out before submission, it creates delays that can kill deals in rate-locked environments.
- Prior government debt. If the borrower has any prior SBA loan defaults, CAIVRS hits, tax liens, or delinquent federal obligations, the deal is dead unless resolved. This needs to be checked before you invest hours in a file.
- The size standard. The SBA defines small business by industry, using either revenue or employee count thresholds. A business that looks small might not qualify as small under its NAICS code. Worth checking, especially in industries with high revenue per employee.
- Change of ownership deals. If the loan is funding a business acquisition, the SBA has specific requirements around valuations, seller notes, equity injections, and standby periods on that seller debt. Getting this wrong is a common and expensive mistake.
What a Strong File Actually Looks Like
Strong files tell a story. They don’t make the underwriter work to understand what’s happening. Every document in the package is there for a reason, and that reason is legible.
Here’s the framework I use when I’m reviewing a file before it goes anywhere:
The Three-Year Picture
Three years of business tax returns plus year-to-date financials. The trend matters more than any single year. If revenues are growing, explain what’s driving it. If there was a dip, explain why and show recovery. If 2020 or 2021 was rough, contextualize it without being defensive.
Lenders understand that businesses have history. What they don’t understand – and won’t give you credit for – is unexplained variance. The story of the numbers is yours to tell. Don’t leave it blank.
The Add-Back Schedule
If your borrower’s tax returns show lower income than what the business actually generates (depreciation, amortization, officer compensation above market, one-time expenses), you need a documented add-back schedule. Not a note in the cover letter. A document, ideally from a CPA, that walks the lender through what actual available cash flow looks like.
This single document can be the difference between a 1.1x DSCR and a 1.4x DSCR. That difference decides the deal.
The Executive Summary
Yes, it still matters. A well-written one-to-two page summary that answers the four questions above, before the underwriter even opens a financial, sets the tone for how the rest of the file gets read. It tells the lender you know what they need and you’ve already thought through the objections.
Most files don’t have one. Or they have a boilerplate version that says nothing. The ones that have a tight, accurate, confident summary move faster and get fewer conditions.
Bank Statements That Match
Twelve months of business bank statements. They should reconcile with what the tax returns say. If they don’t, you need to explain why before anyone asks. Unexplained deposits, large cash transfers, irregular patterns – every one of these generates a condition that slows your deal and raises the temperature in the room.
The Lender Selection Problem
Here’s something most referral partners don’t think about enough: lender selection is part of the underwriting strategy.
Not every SBA lender looks at the same deal the same way. Community banks may favor local operators with real estate. National SBA lenders may be more comfortable with service businesses and intellectual property. Some lenders are aggressive on hospitality. Others won’t touch it at any price.
Sending the right file to the wrong lender is almost as bad as sending the wrong file. You get a decline that becomes part of the paper trail, and now the next lender sees a prior decline and wants to know why.
The credit culture of the lender needs to match the profile of the borrower. This is not something you can know from their website. It comes from relationships, from volume, from having worked deals through multiple institutions and understanding how each one thinks.
The Real Cost of a Messy File
Let’s be direct about what’s at stake. An SBA decline isn’t just a lost deal. It’s a CAIVRS record if there’s default history involved. It’s a lender notation that becomes a question in the next submission. It’s time – weeks, sometimes months – that a business owner spent waiting instead of operating.
And it’s a relationship cost. When a referral partner submits a deal that falls apart in underwriting, that reflects on everyone in the chain. Lenders notice patterns in the submissions they receive. The partners who bring clean, well-packaged files get calls returned faster, get conditions handled more efficiently, and occasionally get a deal done that a messier shop couldn’t land.
The SBA process is not forgiving of sloppiness. It is, however, very forgiving of complexity, when complexity is properly explained and documented.
That’s the whole game. Complexity explained is strength. Complexity unexplained is a decline.
What to Do When the File Isn’t Ready
Sometimes a deal comes in and the file isn’t there yet. Tax returns are late. Books are a mess. There’s a prior MCA stack creating a debt coverage problem. The borrower hasn’t filed in two years.
Don’t force it.
A deal submitted six months early and declined is harder to rescue than a deal that waited six months, got the books cleaned up, and came in clean. Patience here is a strategy, not a delay.
The best thing you can do for a client who isn’t ready is tell them exactly what needs to happen and give them a real timeline. That’s the mentor move. It costs you nothing and builds more loyalty than a quick shot that doesn’t land.
If you have a deal that doesn’t quite fit the standard box – ownership complexity, non-traditional income, messy history with a good story behind it – let’s talk. That’s exactly the kind of file we work best with.
