How a consultant converts trade-area demand, daypart capture, and a defensible cost structure into a lender-grade projection for SBA, USDA, and conventional restaurant financing — in a sector where 42% of operators were unprofitable last year and eat-in dining has lost a third of its share since 2019.
Why restaurants are the hardest common asset class to underwrite
Restaurants are the most-financed category in SBA lending. Full-service and limited-service establishments together account for more than 17,000 SBA 7(a) originations over the last five fiscal years — the largest single industry concentration in the programme. Accommodation and food service as a whole represents roughly 12.5% of all 7(a) loans and $17.9 billion of approvals.
They are also, on the published evidence, not the riskiest.
On seasoned originations, restaurant charge-off rates run around 7.1% — below construction at roughly 7.8% and below retail trade at roughly 7.6%. The sector every credit committee treats as the archetypal high-risk loan sits mid-table, not at the top.
Both facts matter to a feasibility consultant, and they pull in the same direction. High volume plus moderate loss experience means lenders will do these deals, and the file that gets approved is the one where the analysis is credible rather than the one where the concept is exciting. The consultant's job is not to argue that restaurants are safe. It is to demonstrate that this restaurant, in this trade area, with this cost structure, generates enough cash to service the proposed debt through a period when most operators are struggling.
That is a harder analytical assignment than it looks, because the sector's economics leave almost no margin for an optimistic assumption.
What the 2026 operating environment does to a projection
A consultant preparing a restaurant projection in 2026 is working against a specific and well-documented backdrop. Ignoring it produces a document a lender will not believe.
Sales are growing; profitability is not. Total restaurant and foodservice sales are projected at $1.55 trillion in 2026, up from $1.4 trillion, but adjusted for inflation, consumer spending is forecast to rise only about 1.3%. Most of the headline growth is menu pricing, not volume.
42% of operators reported their restaurant was not profitable in 2025. Sixty per cent reported a year-over-year decline in customer traffic. Only 15% said conditions had improved on the prior year.
Costs have reset permanently. Food costs sit roughly 38% above 2019 levels and labour costs roughly 35% above. More than nine in ten operators cite food, labour, insurance, energy and card processing fees as significant challenges. Around two-thirds reported that tariffs on imported food and beverage drove costs higher.
Pricing power is largely exhausted. Around 90% of full-service operators and 85% of limited-service operators raised menu prices last year. The lever has been pulled. A projection that assumes further price increases will carry margin is assuming something the industry has already spent.
Margins are thin even in good conditions. Median pre-tax profit margins run around 2.8% of sales for full-service concepts and around 4% for quick service.
Labour is structurally tight. Full-service employment remains roughly 204,000 jobs — about 3.6% — below pre-pandemic levels. Turnover runs 35% to 40% annually and replacing a single manager costs upwards of $10,000.
For the consultant, the practical consequence is that the projection has to be built to survive scrutiny at these cost levels, not at historical ones. A model using pre-2020 prime cost ratios or 2019 labour percentages is not conservative; it is wrong. Lenders active in this sector know the current numbers, and a projection that shows a 62% prime cost in a market where operators are running 66% invites the question the sponsor least wants asked.
The structural shift that changes what a concept is worth
One data point reframes restaurant feasibility more than any other.
Eat-in occasions as a share of US foodservice dollar sales fell from 56% in 2011 to 51% in 2019, and to roughly 35% by 2025.
That is not a cyclical dip. It is a permanent reallocation of demand from dining rooms to off-premise channels — delivery, takeaway, drive-thru, catering, and pickup. And it has direct consequences for how a consultant sizes and values a concept.
Full-service concepts with large footprints are carrying square footage against a shrinking channel. Black Box data shows 9% of tracked full-service units lost 30% or more of their peak sales between 2019 and 2025, against 4% in limited service. Among the most distressed, 3% of full-service and 1% of limited-service locations lost more than half their peak sales.
The correction is bifurcated, not uniform. Value-oriented formats are capturing share. Mid-market casual dining is under pressure from both ends of the price spectrum — squeezed by fast-casual below and by occasion-driven dining above.
Off-premise capability is now a structural variable, not a feature. A concept's ability to serve off-premise demand determines what share of its trade area's spending it can reach. A consultant analysing a full-service concept without addressing off-premise capacity — kitchen throughput for delivery, pickup staging, drive-thru feasibility, third-party platform economics — has analysed the 2019 market.
This is where a competent restaurant feasibility study separates itself. Anyone can find the trade area population. The analytical work is establishing what share of that population's food-away-from-home spending the subject can actually reach given its format, footprint and channel capability.
What the consultant actually does
Defines the trade area by behaviour, not radius
Restaurant trade areas are drawn by drive time and by daypart, and the two differ.
A breakfast and lunch daypart draws from where people work and commute. Dinner draws from where they live. A destination concept draws further than a convenience concept. A drive-thru captures traffic that would never park.
The consultant establishes the primary trade area — typically five to ten minutes drive time for convenience formats, fifteen to twenty for casual dining, longer for destination concepts — and a secondary area contributing incremental volume. Where the site depends on a specific generator (an interstate interchange, a hospital campus, a university, an employment centre), that generator is analysed separately rather than folded into population counts.
Traffic counts matter and are frequently misused. A high count on an adjacent road is not demand unless the site is accessible from it, visible in time to react, and positioned on the correct side for the relevant daypart. Access, egress, signalisation and turning movements are part of the demand analysis, not the site description.
Sizes demand from spending, not from population
The credible chain runs: trade area households → household income → food-away-from-home spending → the share addressable by this concept's price point and format → competitive share.
Each step needs external evidence. Household counts and income distribution from census data. Food-away-from-home spending from consumer expenditure data, adjusted for local income. Addressable share determined by whether the concept's average check fits the market's spending capacity.
Price point discipline is where many studies fail. A $34 average check in a trade area whose median household income supports a $19 check is not an ambitious plan; it is a mismatch. The consultant's job is to say so before the loan is made rather than after.
Maps competition honestly, including what is closing
Competitive analysis for restaurants means every establishment in the trade area competing for the same occasions — not merely the same cuisine. A fast-casual bowl concept competes with the sandwich shop, the taqueria and the grocery hot bar, not only with other bowl concepts.
For each competitor the consultant establishes format, price point, capacity, apparent volume, and daypart strength. Where possible, this is supported by observation rather than by listing.
Recent closures are evidence. In a period of documented market correction, a trade area that has lost two restaurants in eighteen months is telling you something about its capacity to support another. A study that lists competitors without noting who has failed has skipped the most informative data available.
The pipeline matters as much as the existing set. A concept opening into a market where two competing formats are under construction is entering a different competitive environment from the one it surveyed.
Builds revenue from covers and check, not from a target
The revenue build is the core of the document, and it must be constructed from operational units rather than from a desired outcome.
For full service: seats, turns per daypart, days of operation, average check by daypart, and the off-premise contribution modelled separately.
For limited service: transactions per daypart, average ticket, and drive-thru versus in-store split where applicable.
Each input requires support. Turns are benchmarked against comparable operations of similar format and price point. Check is supported by the proposed menu at proposed prices, not by an industry average. Seasonality is applied where the market is seasonal — a lake-adjacent concept and a downtown office-district concept have opposite seasonal profiles, and an annual average conceals both.
Ramp is modelled explicitly. Restaurants do not open at stabilised volume. Nor do they always open below it — a well-marketed opening frequently produces an initial spike followed by a decline before settling. Lenders have seen enough openings to know that a straight-line ramp to stabilisation is not how it works, and a model showing one signals inexperience.
Builds the cost structure to current reality
This is where credibility is won or lost, because a lender's analyst will check these ratios first.
Prime cost — cost of goods plus labour — is the metric the industry manages against and the one a lender will test. Full service typically targets the low-to-mid 60s as a percentage of sales; limited service somewhat lower. A projection showing materially better prime cost than comparable operations needs an explanation, and "efficient management" is not one.
Food cost is built from the proposed menu with current input pricing, not from a percentage assumption. In an environment where food costs sit 38% above 2019 and tariff effects have been absorbed rather than reversed, using a historical percentage understates.
Labour is built from a staffing schedule by daypart at local prevailing wages, including payroll taxes, benefits where offered, and — critically — the cost of turnover. At 35% to 40% annual turnover and $10,000-plus to replace a manager, recruitment and training are recurring operating costs rather than one-time opening expenses.
Occupancy at contracted rent plus triple-net charges, with escalations modelled across the projection period.
The lines operators consistently understate: insurance, which has risen sharply; card processing fees, which scale directly with sales; repairs and maintenance, which arrive unevenly and are frequently omitted entirely in year one; and marketing, which cannot be zero for a new concept.
Replacement reserves. Kitchen equipment, HVAC, furniture and smallwares have finite lives. Where the financing is USDA-guaranteed this is not optional — 7 CFR Part 5001 defines debt service coverage as EBITDA less reasonably expected replacement capital expenditures over annual debt service, so an unfunded reserve directly reduces the coverage ratio the lender computes.
Tests the structure, not just the base case
A single stabilised-year coverage figure tells a credit committee very little. The consultant's sensitivity work should establish:
Coverage at 90%, 85% and 80% of projected revenue, and specifically where coverage crosses 1.00x
What a 200 basis point move in prime cost does to coverage
What a delayed opening does to the ramp and to the interest reserve
For leased premises, what rent escalation does across the loan term
Where the concept depends on a single daypart or a single generator, what happens if that source weakens
The purpose is not to prove the project survives everything. It is to identify the variables that actually govern the outcome and quantify them, so the lender is making a priced decision rather than a hopeful one.
Assesses management as a credit factor
In a sector with 2.8% median margins, operator capability is not a soft factor.
The consultant addresses whether the sponsor has operated this format at this scale, who runs the kitchen and the floor, what the compensation structure is, and what happens to the business if the key operator leaves. For a first-time operator — and roughly a quarter of full-service SBA borrowers and 40% of limited-service borrowers are financing a startup — this section carries disproportionate weight and needs to be answered directly rather than glossed.
How the programmes differ
SBA 7(a) and 504. Restaurants are heavily financed under 7(a), with average approvals around $574,000 for full-service and $501,000 for limited-service. Where the transaction includes real estate, 504 may apply. SOP 50 10 8 sets the circumstances in which a third-party feasibility study is expected — startups and businesses under two years old, changes of ownership, new construction or major expansion, and special-purpose property. A restaurant acquisition by a first-time operator hits two of those simultaneously.
USDA B&I. Available for restaurants in communities of 50,000 or fewer, up to $25 million, with fiscal 2026 guarantees at 85% below $5 million and 80% at or above. USDA's feasibility scope under Part 5001 Appendix A to Subpart D adds an economic-impact dimension that SBA does not require: what the project does for the rural community, in jobs and in economic contribution. That section is scored, not decorative.
Conventional, including bank and credit union. No prescribed scope, but lenders frequently require a feasibility study on restaurant credits precisely because the sector's failure rate is well known to them. The absence of a regulatory template means the consultant is writing to the lender's credit policy, and asking for that policy in advance is worth doing.
Across all three, the study must be prepared by an independent third party with no financial interest in the transaction. A projection prepared by the sponsor, the broker, the franchisor or the equipment supplier does not satisfy the requirement and, more practically, does not persuade.
Franchise restaurants: a different analysis
Roughly one in eight SBA loans goes to a franchised business, and restaurants are heavily represented.
A franchise restaurant feasibility study is not a shorter version of an independent one. It has additional obligations:
Item 19 financial performance representations, where the franchisor provides them, are the single most useful comparable data available — and their absence is itself informative.
Unit-level economics from the system, including how the subject's proposed trade area compares to the system average on population, income and competitive density.
Closure and transfer data. A system's unit count growth conceals churn. Openings minus closures is the number that matters.
Royalty and marketing fund obligations modelled as the fixed percentage costs they are, before any operator-level margin.
Territory and encroachment terms, which determine whether the trade area analysed today remains the trade area in five years.
The critical discipline is that brand recognition is not demand. A known brand in an oversupplied trade area still has to capture spending that is already committed elsewhere.
What a lender is actually reading for
A restaurant feasibility study is read by someone who has seen dozens fail. They are looking for four things.
Is the revenue reachable? Not whether the concept is good, but whether the covers and check assumed are achievable in this specific trade area given the competition and the format.
Is the cost structure real? Prime cost, labour schedule and the omitted lines. This is checked first and fastest.
Does it cover, and by how much, and under what conditions? Coverage at stabilisation, coverage through the ramp, and coverage under stress.
Can this operator execute it? In a 2.8% margin business, the answer to this question frequently determines the credit.
A study that answers those four clearly, with sourcing a reviewer can check, does its job. A study that surveys the market attractively without concluding does not — and the concluding opinion is, in most programme contexts, a formal requirement rather than a stylistic preference.
The honest position
Restaurants are financeable. The data says so: they are the most-financed category in SBA lending and their loss experience sits below construction and retail. Lenders will do these deals.
But the sector is in a documented correction. Sales growth is nominal. Traffic is soft. Costs have reset upward permanently. Pricing power is spent. Eat-in dining has surrendered a third of its share since 2019, and the correction is hitting large-footprint full-service concepts hardest.
In that environment, the value of an independent feasibility study is not advocacy. It is that a lender receives a projection built from evidence rather than optimism, with the downside identified and quantified — and that a sponsor finds out before signing whether the concept, the site and the capital structure actually fit together.
Sometimes the most valuable thing a consultant delivers is the finding that they do not.
Prepared by feasibility-study-consultant.com. Industry data reflects published sources at the date of writing. Programme requirements should be verified against current SBA and USDA guidance. Last updated: July 30, 2026.