Why Feasibility Studies Fail Lender Review: A Taxonomy of Rejection

A feasibility study fails lender review when it is built to sell a deal rather than to evaluate one. Regulated lending files are not read once and filed. They are read by underwriters, by credit committees, by agency reviewers and bank examiners, and — if the loan later defaults — by counsel and the SBA Office of Inspector General. Every one of those readers is checking the document against an external standard, and when the study cannot be reconciled to that standard, rejection is not bad luck. It is the rational outcome.

The reassuring part, for a sponsor, is that the failure is almost never the business idea. It is that the document does not answer the questions the governing regime requires it to answer — and those questions are knowable in advance. Nearly every rejection traces to a specific rule: a provision of SBA's SOP 50 10 8, a section of USDA's 7 CFR Part 5001, or an appraisal and underwriting convention such as USPAP Standards Rule 1-3 or the debt-service-coverage floor. What follows is a taxonomy — each failure mode, what it looks like in a returned report, and the rule it violates.

The three standards every study is graded against

Before cataloguing the failures, separate the standards a reviewer applies, because confusing them is itself a source of failure. They fall into three tiers, and a report that treats a soft convention as optional, or a hard requirement as a convention, has misread the assignment before the first number is run.

The first tier is the hard regulatory requirement — the codified must-haves in SBA's SOP 50 10 8 and USDA's 7 CFR Part 5001. For the loans they govern, these are binding, and a study that omits what they require is deficient on its face. The second tier is appraisal and analytical standards: the Uniform Standards of Professional Appraisal Practice (USPAP), particularly Standards Rule 1-3 on market and highest-and-best-use analysis; the Appraisal Institute's market-analysis framework; and asset-specific conventions such as STR's competitive-set rules for hotels and NCHMA's Model Content Standards for multifamily. These are not federal statutes, but for the loans and asset classes they govern they function as binding. The third tier is softer underwriting convention: DSCR floors, expense-ratio benchmarks, reserve sizing, and the general expectation of conservatism a credit committee applies as a matter of prudence.

Conventional lenders operate under supervisory guidance rather than a codified feasibility mandate, but construction lenders, CMBS conduits, and life-insurance lenders routinely require independent demand and financial analysis for hospitality, healthcare, and other capital-intensive projects. USDA's own §5001.202 credit-evaluation standard — requiring analysis against "industry standards (such as Dun & Bradstreet or the Risk Management Association)" and documentation of any projection "deviating from historical financial performance" — is a fair proxy for what any conventional or institutional credit committee expects.

SBA 7(a) and 504: a discretionary rule, a non-negotiable expectation

The governing edition of SBA's operating procedure is SOP 50 10 8, effective June 1, 2025, which replaced SOP 50 10 7.1 and carries a technical update effective March 1, 2026. It is the operative manual through mid-2026. Read literally, its feasibility hook is discretionary: the SBA may require a study, and it sits in the rulebook beside appraisals and surveys as a tool the lender can invoke rather than a universal mandate. In practice, the discretion is narrow. Lenders and SBA quality reviewers expect a study wherever projections, not operating history, carry the credit — startups, ground-up construction or major expansion, complete changes of ownership, and, above all, special-purpose property.

The SOP defines special-purpose property as "a limited market property with a unique physical design, special construction materials, or a layout that restricts its utility to the specific use for which it was built," and its examples read like a list of the busiest independent-feasibility asset classes: hotels, car washes, gas stations, and assisted-living facilities among them. For special-purpose property, SBA treats the appraisal and the feasibility study as separate deliverables answering separate tests — the appraisal supporting the collateral test, the feasibility study the cash-flow test.

Where a going concern is involved, the SOP requires the appraisal be performed by a Certified General Real Property Appraiser who has completed no fewer than four going-concern appraisals of equivalent special-use property within the last 36 months, with value allocated separately to land, building, equipment, and intangible assets. On repayment, a standard 7(a) loan must show a DSCR of at least 1.15x, with operating cash flow defined as EBITDA; 7(a) small loans of $350,000 or less must clear 1.10x under the March 1, 2026 update; and a global cash-flow analysis is required wherever affiliates are involved. Miss any of these and the credit is not merely weak. It is non-conforming.

USDA's 7 CFR Part 5001: the most explicit mandate in federal lending

If SBA's requirement is discretionary, USDA's is the opposite — the most explicit and quotable feasibility mandate in federal lending. 7 CFR Part 5001, the codified home of the OneRD Guaranteed Loan Initiative, took effect October 1, 2020, consolidating six legacy regimes, including the old Business and Industry (B&I) rules at 7 CFR Part 4279 Subpart B. That consolidation matters, because most online guidance about "B&I feasibility requirements" still cites regulations that no longer govern current origination. Section 5001.3 defines a feasibility study as "a report including an opinion or finding conducted by an independent qualified consultant(s) evaluating the economic, market, technical, financial, and management feasibility of the proposed project or operation in terms of its expectation for success as outlined in appendix A to subpart D of this part."

The trigger is mechanical, not judgmental. Under 7 CFR 5001.306(a)(3)(i), a feasibility study by an independent qualified consultant acceptable to the Agency is mandatory for any B&I guaranteed loan greater than $1,000,000 to a new business — itself a regulatory term of art, defined in §5001.3 as a business in operation less than one full year, or one operating at least a year but not yet at full operational capacity or stable operations. For loans of $1,000,000 or less, the Agency may require a study under §5001.304(a)(4)/(ii) where the lender's own analysis cannot establish technical feasibility, market feasibility, or economic viability, or where the project significantly affects an existing borrower's historic cash flow. Community Facilities financing splits under §5001.304 into a lighter financial feasibility analysis — available under three safe harbors, including loans of $25 million or less to existing facilities — and a heavier financial feasibility study with examination opinion, prepared to AICPA attestation standards by a consultant carrying professional liability insurance. For REAP, §5001.307 scales the technical report to project cost and reserves a full study for cases the lender or Agency deems necessary.

The analytical spine of every USDA study is the set of five components codified in Appendix A to Subpart D, and they are worth stating in the regulation's own words, because reviewers check for them by name. The economic component requires a "cost benefit analysis" weighing the "minimum amount of inputs (labor, infrastructure, utilities, renewable resources, feedstocks) to operate successfully," contracts, environmental risks, and the "overall economic impact of project including new markets created and economic development." The market component requires "analysis of the current and future market potential, competition, sales or service estimations including current and prospective buyers or users." The technical component requires "analyzing the reliability of the technology to be used and/or the analysis of the delivery of goods or services, including transportation, business location, and the need for technology, materials, and labor." The financial component requires "analysis of the operation to achieve sufficient income, credit, and cashflow to financially sustain the project over the long term and meet all debt obligations," and explicitly names "sensitivity analysis" and "peer industry comparison" among its factors. The management component requires "analysis of the legal structure of the business or operation; ownership, board and management analysis." The appendix also requires an executive summary, a consultant's recommendation, and a statement of the author's qualifications. All five are mandatory; a study omitting any one is facially incomplete.

Why the reviewer is reading harder than before

None of this scrutiny is theoretical, and it has intensified. USDA's own approval data shows the share of B&I applications receiving approval falling from 89.1 percent in FY2021 to 52.7 percent in FY2023 — the most recent year for which throughput data has been published, as reported by MMCG Invest drawing on Office of the Comptroller of the Currency figures. Nearly half of applications now fail, and the study is the first document a reviewer attacks.

The portfolio behind that tightening explains it. On February 20, 2026, USDA RBCS Administrator J.R. Claeys sent an open letter to more than 775 OneRD lenders disclosing an active guaranteed balance in excess of twelve billion dollars, delinquent loans in excess of one billion, and roughly $300 million in repurchases and losses over the prior twelve months — and reminding lenders of the Agency's removal authority under 7 CFR 5001.132. When an agency is absorbing losses at that scale, it reads the feasibility study the way a claims adjuster reads an application: looking for the reason to say no. A study written to survive that reading is a different document from one written to close a deal.

Market and demand failures: the most common and most fatal

The single most common failure category, and the most fatal, is unsupported market demand. It shows up as demand assumptions with no evidentiary floor; weak or absent trade-area delineation; a primary market area defined by radius when it should be drive-time, or defaulted to a county or place boundary; national averages standing in for local data; competitive supply mishandled or omitted; and capture or penetration rates asserted rather than derived.

The rules this violates are specific. For USDA, market feasibility is one of the five codified Appendix A components, so a study that asserts demand without a supply collision fails on the regulation's own terms. For any appraisal-adjacent study, USPAP Standards Rule 1-3(a) requires the analyst, in a market-value context, to "identify and analyze the effect on use and value of … economic supply and demand … and market area trends," and its comment is blunt: "an appraiser must avoid making an unsupported assumption or premise about market area trends." The Appraisal Institute's six-step market-analysis process — property productivity, market delineation, demand analysis, competitive supply analysis, residual demand, and capture analysis — is the recognized method, and skipping the supply and residual-demand steps is a methodology failure, not a shortcut.

The asset-specific conventions are just as concrete. For hotels, the STR competitive-set guidelines, effective January 1, 2017, require a minimum of four properties excluding the subject, at least three of them unaffiliated, representing at least two unaffiliated companies, with no single brand exceeding 50 percent and no single company exceeding 70 percent of participating room supply; a compset that violates those ratios is not defensible. For multifamily and LIHTC, NCHMA's Model Content Standards, updated September 2025, govern capture- and penetration-rate methodology — treating a capture rate below ten percent as generally acceptable and up to fifteen percent acceptable in documented fast-growth markets, and requiring capture to be run by AMI band and bedroom count rather than in aggregate.

Financial projections and the DSCR floor

The second category is aggressive financial projection: a revenue ramp or absorption assumption that outruns the evidence, expense ratios inconsistent with industry benchmarks, break-even and sensitivity analysis missing, and — the recurring offender — a model that books stabilized performance from day one and ignores the ramp. The mechanics of deriving a defensible ramp are the subject of a companion note on absorption forensics.

The hardest number in the file is the DSCR floor, and it is where most financial rejections happen. Under SOP 50 10 8, the SBA requires a minimum projected DSCR of 1.15x — 1.10x for 7(a) small loans under the March 1, 2026 update — while most SBA lenders target 1.25x internally and the most conservative require 1.50x. Conventional floors are tiered rather than singular. Freddie Mac's Optigo Small Balance Loan program, for instance, ranges from a 1.20x minimum in top markets to 1.50x for full-term interest-only financing in very small markets. For conventional multifamily agency debt, both GSEs sit at 1.25x: Freddie Mac's Multi PC Program Overview, as of March 31, 2026, states that "for fixed-rate Conventional Loans, the debt-service coverage ratio (DSCR) must be at least 1.25x for the first mortgage," and Fannie Mae's Standard DUS product carries "a minimum 1.25x DSCR requirement."

A conclusion of feasibility that rests on a DSCR dipping below the lender's floor in any projection year — even one — will be returned. USDA reinforces the point from another direction: §5001.202 requires that "financial projections deviating from historical financial performance must be substantiated and documented" and that any increase to revenues and margins "be reasonable and substantiated," and Appendix A lists sensitivity analysis as a required financial factor, so its absence is a codified deficiency rather than a stylistic one. The most common structural trick in this category is annual averaging that hides a sub-1.0x quarter; lenders who pull quarterly statements re-cast the model to the worst quarter, and a study that averages its way over the floor is inviting exactly that recast.

Independence, and the "study" that is really a business plan

The third category is independence and qualification: a preparer who is not independent, lacks asset-class credentials, or carries a conflict of interest — or a "study" that is really the borrower's own projections or business plan under a new cover. USDA is again the most explicit. Section 5001.3 requires an "independent qualified consultant(s)," and §5001.306(a)(3)(i) requires the B&I preparer be "acceptable to the Agency." A consultant who also brokers, develops, or arranges financing for the same project is not independent.

The key distinction, often misunderstood, is that independence attaches to the absence of a stake in the outcome, not to who pays the fee: a borrower-paid study is perfectly acceptable so long as the compensation does not depend on the conclusion or on loan approval. For SBA special-purpose going-concern appraisals, SOP 50 10 8 requires independence from the loan-production and approval functions and prohibits using a valuation prepared for the borrower or seller. And USPAP's Ethics Rule — which reaches market studies performed as "other" valuation services under Advisory Opinion 21 — prohibits the analyst from advocating a party's cause. A business plan is an advocacy document written by the borrower to raise capital; it cannot substitute for an independently prepared feasibility analysis, and a reviewer recognizes the difference immediately.

Data, sourcing, and methodology

Two related categories account for a large share of returns that have nothing to do with the strength of the market: stale or unsourced data, and opaque methodology. On data, the failures are familiar — outdated Census or industry figures, national averages standing in for a rural submarket, no primary research, no citations. USDA §5001.202 requires comparison against current industry standards such as the Risk Management Association's Annual Statement Studies or Dun & Bradstreet, and USPAP Standards Rule 1-1(b)–(c) prohibits substantial errors of omission or commission and careless or negligent analysis. In thin-data rural markets, reviewers do not expect national data; they expect creative but documented sourcing — ACS five-year estimates, USDA Economic Research Service data, state economic-development reports, and genuine primary research — precisely because national averages cannot describe a rural submarket. An unsourced claim reads as a report designed to validate a conclusion rather than to test one.

On methodology, the failure is the black box: a conclusion the reader cannot reproduce. USPAP Standards Rule 1-1(a) requires the analyst to "correctly employ those recognized methods and techniques necessary to produce credible results," and the Appraisal Institute's six-step process is the recognized method for market analysis — a conclusion that cannot be traced through demand, supply, residual demand, and capture is not reproducible. A study can present an abundance of accurate data and still fail the credibility test if it never demonstrates the data's relevance or draws a conclusion from it. Our own analytical framework is built to make every conclusion traceable back through those steps.

Scope and presentation: the avoidable returns

The last two categories are the most avoidable, which is what makes them the most frustrating when they trigger a return. Scope failures occur when a study, however competent, does not answer the specific questions the program requires. For USDA, all five Appendix A components are mandatory, so a study that omits management feasibility or economic impact is incomplete regardless of how thorough its market section is. For SBA special-purpose property, the analysis must align with the SOP's separate cash-flow test. The standard a conventional lender applies differs from an SBA lender's, which differs again from a USDA transaction or an institutional investor's, and a study that is technically sound but built for the wrong audience can be returned simply for leaving the actual credit decision unanswered.

Presentation and completeness failures are the quiet killers: internal inconsistencies, math errors, conclusions the body does not support, missing certifications. A net-profit figure in the P&L that does not tie to the cash-flow statement is one of the most common reasons a credit officer returns a report, because officers are trained to cross-check figures and a single inconsistency undermines confidence in every number in the file. USDA Appendix A requires a statement of the author's qualifications and a consultant's recommendation, and their absence is a codified gap; USPAP Standards Rule 2-3 makes a signed certification an integral part of the report. None of these defects reflects a weak project. All of them get a study returned.

The three studies people conflate

A surprising share of scope failures come down to a vocabulary problem: conflating a market study, a marketability study, and a feasibility study, which are three different instruments. A market study, in the Appraisal Institute's Dictionary of Real Estate Appraisal, is "the study of the supply and demand in a specific area for a specific type of property" — macro conditions. A marketability study is site- and property-specific and follows the six-step process to the subject's own capture. A feasibility study is a cost-benefit comparison — "a comparison of the cost versus the value of a project" — that asks the forward question the appraisal does not: given this market, this competition, and this cost structure, will the project generate the cash flow to service the loan through the ramp and at stabilization?

The practical consequence is that an appraisal is a present-value opinion supporting the collateral test, while a feasibility study supports the cash-flow test, and the federal frameworks do not permit one to substitute for the other — many loans require both. Delivering a market study when the program requires a feasibility study is a scope failure that guarantees a return, no matter how good the market study is.

Highest and best use, the hidden test

One appraisal doctrine sits underneath the whole exercise and quietly fails studies that ignore it. The Appraisal Institute defines highest and best use as "the reasonably probable and legal use of vacant land or an improved property that is physically possible, appropriately supported, financially feasible, and that results in the highest value," tested against four criteria: legally permissible, physically possible, financially feasible, and maximally productive.

The third criterion is the trap. Financial feasibility — generally, a use whose net present value exceeds zero — cannot be established without a supply-and-demand analysis, which is precisely what a feasibility study exists to provide. A study that ignores the highest-and-best-use framework can fail an appraisal-informed review because the "financially feasible" prong is exactly the question the study was commissioned to answer, and USPAP Standards Rule 1-3(b) requires the analyst to "develop an opinion of the highest and best use of the real estate" whenever the assignment involves market value. A feasibility conclusion that cannot be squared with highest and best use has an unresolved contradiction at its center.

Writing a study that survives the review

The remedy for all seven categories is structural, not cosmetic, and it runs in four stages. Before commissioning, identify the governing regime first, because the regime dictates required scope, preparer credentials, and whether a study is even mandatory. A USDA B&I loan over $1,000,000 to a new business requires a full five-component study, not an optional one; an SBA special-purpose credit needs an analysis aligned to the cash-flow test; a conventional deal answers to its lender's underwriting file. Confirm at this stage that the preparer is genuinely independent — no brokerage, development, or financing interest in the project — and demonstrably competent in the asset class.

During preparation, insist on the things reviewers check: a drive-time-validated primary market area; a competitive set built to the applicable standard, STR for hotels and NCHMA for multifamily; a forward supply pipeline that includes shadow supply; capture and penetration framed by segment; ramp-year modeling on as-is expense ratios with a sized operating reserve, commonly twelve to eighteen months of debt service for ground-up construction; and a sensitivity analysis that states what the project must clear, not what the analyst hopes it will. Every material claim carries a named, dated source.

Before submission, run the file against the review the lender will run. Does the projected DSCR clear the floor in every year, including the worst quarter? Are all five USDA components present? Do the P&L, balance sheet, and cash flow reconcile? Is the methodology stated and reproducible? Are the certification and qualifications statements attached? If the projected DSCR falls below 1.15x for SBA, or the lender's stated conventional floor in any year, the deal is restructured — more equity, longer amortization, a smaller loan — not re-argued. And on revision, treat a return as a scope conversation, not an editing pass: map each reviewer comment to the specific provision it implicates and remediate at that level. A study is not finished when the analyst is satisfied. It is finished when the reviewer accepts it — and the surest way there is to write, from the first page, the document the reviewer was going to demand anyway.