Gas Station and C-Store Feasibility and Market Studies

NACS's April 2026 release is the broadest public benchmark: in-store sales of $341.2 billion in 2025, up 1.7%, and fuel sales of $476.3 billion, down 5.4% as pump prices fell while gallons rose 0.5% ³. The full survey is sold commercially, and its public summary gives no average operating expense ratio, NOI margin or payroll share, so our studies take those ratios from operator filings and label them.

Fuel margin: three measures, one cycle

NACS reports a gross gasoline margin of 39.7 cents a gallon in 2024, 11.9% of a $3.33 average price, and a four-year average of 38.0 cents ⁶; its research director said margins averaged over 40 cents in 2025 ⁴. OPIS data compiled by a bank give five-year averages of 39.3 cents for gasoline and 54.3 cents for diesel ²³. The chains' own second-quarter 2026 margins ran from 35.1 cents at Murphy USA to 52.6 cents at Couche-Tard's U.S. business and at Yesway, against 29.2 to 44.9 cents a year earlier ²³, with crude at $97.26 a barrel on September 9, 2026 and the New York Harbor gasoline crack spread about $1 a gallon above 2025 since May ²³. The base case belongs near the multi-year averages, adjusted for the subject's brand and competitors. The downside uses the prior year's low. Margin above the averages is upside, not debt capacity.

Volume and inside sales

The average store sells about 4,000 gallons of gasoline a day ⁶, but format decides the spread: Yesway's new large stores averaged 3.1 million gallons a year against 1.0 million at its legacy stores, and earned $1.4 million of store contribution against $0.4 million ⁷. Same-store gallons in the second quarter of 2026 ran from minus 8.1% at CrossAmerica to plus 1.4% at Yesway ²³. Gallons come from the subject's trade area, as the demand section describes, checked against the format it most resembles. Inside, foodservice was 28.5% of in-store sales and 38.9% of in-store gross profit in 2025 ³, yet transactions per store fell 2.7% ³, fuel transactions reached their lowest since 2021 ⁴, and the average inside transaction lost 7 cents after expenses despite a 24-cent rise in basket value ⁵. An industry-average inside margin applied to projected transactions, without card fees, shrink and food labor, shows a profit the industry did not earn that year.

Labor and operating expense

The industry averaged 19.9 employees per store at $15.04 an hour in 2025 ³. Yesway runs about 2.6 employees per shift, and its salaries and benefits came to 7.5% of revenue ⁷, a ratio that swings with the pump price, so labor should be built from staffed hours and local wages. Card fees reached a record $21.3 billion and total expenses are up 23.3% since 2021 ³⁴; card fees scale with the pump price and need their own line. Murphy USA's store operating expense excluding payment fees and rent, $36,200 to $36,600 per store per month ¹², is a floor for a lean, high-volume format. Third-party management fee norms and franchise Item 19 figures were not found, so the study takes the subject's actual agreements.

The governing text is SOP 50 10 8.1 with Technical Policy Updates, effective October 1, 2026 ³¹. It names gas stations among its examples of special purpose property, defined as a limited-market property with a unique physical design, special construction materials, or a layout that restricts its utility to the use for which it was built ³¹. On the study the regulation is unchanged: 13 CFR 120.160(b): SBA may require professional appraisals, a survey, or a feasibility study ³². Lenders ask for one on a fuel site when there is no operating history to underwrite. Our SBA 7(a) and 504 practice area covers the program rules in full.

7(a) for a new store

A new store financed with 7(a) needs 10% of total project cost as equity, and its projections must reach 1.15x coverage within two years ³¹. The loan can reach $5 million, with real estate amortized over up to 25 years. Year two carries the file: no published benchmark gives a ramp for new convenience stores, so the ramp is argued from the subject's competitors and the operator's own record.

7(a) for an acquisition

Equity is 10% of total project cost and cannot be waived, and coverage of 1.25x is measured on historical EBITDA over total post-transaction debt service, on the last fiscal year-end or the average of the last two ³¹. For most businesses projections are reviewed but cannot meet the floor, except an owner-occupied Special Purpose Property acquisition whose price the appraised value fully supports, where the floor may be met on projections within two years ³¹. That exception reaches gas stations directly and puts the appraisal and the study at the center of the file. Quality of earnings work is not required for Owner Buyout, ESOP & Cooperative, and any acquisition of an owner-occupied Special Purpose Property. The whole loan can run 25 years when 85% or more of the use of project cost is real estate and the station is integral to the business; otherwise the business portion amortizes over 10 years ³¹. The Opelika acquisition case works the historical test through.

504 and the combined limit

The 504 borrower contributes 10% in the standard case, 15% for a new business or a special purpose property, and 20% for both ³¹, so no fuel project sits in the bottom tier and a new operator building a station sits in the top one. The structure follows the contribution: Third Party Lender 50%, debenture 40%, borrower 10%; debenture 35% at a 15% contribution and 30% at 20%. Coverage is 1.15x, the debenture is capped at $5 million, and since July 4, 2026 a 7(a) loan followed by a 504 loan can reach $10 million combined ²⁷³¹, which brings larger interchange formats within reach.

Fees, USDA and banks

For loans approved from October 1, 2026 the upfront 7(a) guaranty fee is 0% on loans of $700,000 or less to businesses in a rural area, meaning a county that the U.S. Census Bureau has defined as at least 30 percent rural ³³. The purchase of an older small-town store can fit under that size, and its study should state the county's rural share. USDA's Business and Industry guarantee is the rural alternative, and its rule differs in kind: for a guaranteed loan over $1,000,000 to a new business it requires an independent feasibility study ⁴². The program page lists an 85% guarantee below $5 million and 80% at or above, a 3% initial fee, a 0.55% annual retention fee and terms up to 40 years ⁴⁰, while September 2025 testimony gave the fiscal 2025 guarantee as 80% ⁴¹, so the percentage should be confirmed for the year of closing. Our USDA Rural Development practice area sets out the study components the rule defines. Conventional terms are the least documented: no lender survey or bank disclosure gives 2026 loan-to-value, coverage or pricing for fuel sites.

Value evidence for fuel sites comes in three tiers that should never be mixed. The first is the credit-tenant net lease, a long corporate lease priced on the tenant's balance sheet. Asking cap rates for 15-year leases in the first quarter of 2026 ran 4.90% to 5.20% for Wawa and 5.35% to 5.65% for Circle K, the 7-Eleven range had tightened to 4.80% to 5.15% by the third quarter, and a 7-Eleven lease with five years left asked 5.90% to 6.20% ⁴³⁴⁴. A listing aggregator found a median Wawa asking rate of 5.00% across 59 listings in October 2026 ⁴⁶, against 6.80% for single-tenant net lease as a whole in the first quarter ⁴⁵. These rates say what an investor pays for a chain's promise to pay rent, not what a store is worth to the person running it.

The second tier is the operator sale-leaseback, where the buyer underwrites the operating company rather than a rated parent: Yesway's build-to-suit program averaged 7.86% across 29 projects ⁷. The third tier, the independent owner-operated station held in fee simple, secures most SBA loans, and no source publishes a cap rate for it, a price per site for single-store sales or a norm for splitting value between real estate and business. The nearest U.S. transaction evidence is a portfolio of 29 Iowa and Kansas stores sold for $17.5 million plus inventory ⁷, a price shaped by the regulatory capital those stores faced.

Operating companies trade on cash flow. Listed convenience operators traded at a median 9.3 times trailing EBITDA on September 10, 2026, below their ten-year median of 10.2 times, with a spread from Casey's at 16.0 times to ARKO at 5.2 times ²³. Couche-Tard's agreement to buy Poland's Żabka, announced on July 31, 2026, was priced at about 9.3 times trailing adjusted EBITDA on an enterprise value near $11.0 billion ²³. A single store carries none of the scale or growth behind those multiples, so they are a ceiling for reasonableness, not a comparable.

For an SBA loan the appraisal does the valuation work. Special purpose property calls for a going-concern appraisal by a Certified General appraiser with recent going-concern experience on equivalent property, allocating value among land, building, equipment and intangibles ³¹. SOP 50 10 8.1 asks for an appraiser with at least four going-concern appraisals of equivalent special use property in the past 36 months, which the engagement letter should require by name ³¹. On an acquisition, total acquisition debt, including seller debt not on full standby, is capped at the business valuation; any excess must be equity ³¹, and where the business purchase price is at or below $350,000 the lender may value the business itself. The study supplies the stabilized income the appraisal capitalizes. The 10-year Treasury yield stood at 4.96% on September 11, 2026 and passed 5% three days later, its highest level since October 2023 ²³.

No federal or lender standard says how many people a gas station needs. Fuel retail has no codified demand ratio, and we found no regional per-capita, fair-share or capture benchmark from a primary source. NACS counts one store for every 2,257 residents on a 343 million population base, or 4.43 stores per 10,000 people ¹. That ratio checks a finished analysis rather than producing one: a trade area with fewer residents per store may be saturated or may sit on a corridor that imports its customers. Our note on gas station demand per capita discusses the ratio's limits.

The method, step by step

1. Draw the trade area on the road network. Drive times, interchange geometry, medians and turning movements decide which cars can reach the pumps; a circle on a map counts customers who cannot turn in. Our note on why the trade area is the study sets out the approach.
2. Size the fuel pool. Start from the vehicles based and commuting in the trade area and from the state transportation department's count stations on the roads that serve the site, naming each station and its count year. In 2025 the average vehicle used 457 gallons, bought 8.4 gallons per fill and came to a station 54.4 times ⁶, which turns a vehicle count into fill-ups.
3. Census the competitors. List every fuel retailer by name, fueling positions, brand, food program and hours, including big-box fuel: 6,862 hypermarket and club sites sell about twice the volume per site of traditional retailers ⁶. Announced and permitted stores belong in the census too.
4. Assign a fair share and a capture rate from fueling positions, adjusted for access, visibility, price position and brand.
5. Build inside sales from transactions. Estimate the fuel customers who come inside, transactions per day and the basket by category, and size foodservice to the kitchen and labor the operator will fund.
6. Cross-check the result against residents per store, the volumes of the format the subject most resembles, and the operator's own record.

A pool that shrinks slowly

U.S. motor gasoline consumption averaged 8.9 million barrels a day in 2025, down 1% from 2024 and 4% from 2019, and EIA forecasts further declines in 2026 and 2027, assuming fleet fuel economy improves about 1% a year while employment grows 0.3% and the working-age population 0.1% a year ⁴⁹. EIA pairs a 1.2% rise in vehicle miles in 2025 with a 1.9% gain in implied fuel economy ⁴⁹, while the highway agency's own year-end count shows miles up 0.9%, or 29.8 billion vehicle-miles ⁷²; a study should cite one series and say which. Either way gallons come from share, not market growth, and a pro forma that grows per-store gallons faster than the trade area adds vehicles needs a stated reason.

Price and season

Price shapes the pro forma more than demand. EIA's October 2026 outlook puts the 2025 retail gasoline price at $3.10 a gallon, reports $4.35 for September 2026, and forecasts $3.91 for 2026 and $3.56 for 2027 ⁵⁰. Higher prices raise card fees and squeeze the inside basket, so the study should state the price its revenue line assumes. National seasonality is modest: demand ran from 8.483 million barrels a day in January 2025 to a peak of 9.262 million in June, and August was 8.8% above January ⁶. A tourist corridor, a lake town or a harvest-season farm market swings far more, and the study should then show monthly volume. No source publishes a monthly index for inside sales.

Bigger stores, more gallons each

Chains that build large stores are raising per-store volume in a flat market: Casey's average gallons sold per store rose from 1,102 in fiscal 2024 to 1,217 in fiscal 2026, in the units its 10-K table defines ⁵¹. That is share moving toward bigger, better-located stores, the same competition a new independent store faces. Where a trade area lacks a modern store, a new one can take share quickly; our note on underserved gas station markets explains how to find those places and tell them from places that are simply small.

Fuel retail is segmented along three lines at once. Format separates a fuel store, a store without fuel and a store built around food. The supply contract, branded or unbranded, applies to every fuel format. Location cuts across both, since an interchange site can hold any format. NATSO treats travel plazas as a channel of their own. The four formats most SBA files involve follow below: branded fuel, unbranded and independent stations, convenience stores without fuel and foodservice-led and QSR stores.

How it earns. An independent buys gasoline without a refiner's flag, at the rack through a jobber or on dealer terms from a distributor, and usually runs one store or a small chain. It can price under the branded store across the road, and it carries the whole risk when wholesale costs jump before the street price follows. No published margin or revenue exists for independents; the nearest evidence is the supplier's side of the trade. ARKO's wholesale business kept 6.3 cents a gallon in 2025 at dealer sites, where the dealer keeps the station margin, and 27.5 cents at consignment-agent sites, where the supplier owns the fuel until it is sold ⁶⁸. Neither is the independent's own margin, but together they show the shape of the deal, so the pro forma has to say which arrangement the borrower has and take its margin from that arrangement's invoices. The inside store does not automatically rescue a thin fuel margin: 70% of stores in the 2024 industry survey sample would have shown a negative in-store operating profit once expenses were charged to the inside business ⁷³.

What it costs. No development cost for an independent station was found, and the franchise ranges in the branded section describe a different product, so a new independent store needs a contractor's bid and its own equipment quotes. On a former station site the larger cost risk is underground: a contractor's summary of EPA's 2022 study put the median cleanup of a former fuel site at $27,120 to $265,883 across three states ⁷¹. For a purchase, the cost evidence is the seller's tank and line tests, leak detection records and state tank registration, which show whether the buyer is acquiring a working fuel system or a replacement project.

How lenders underwrite it. SBA's special purpose list names gas stations without distinguishing branded from unbranded ³¹, so the collateral class is the same. Most independent files are acquisitions under the change-of-ownership rules in the capital section, and seller debt on full standby, other full-standby debt and minority investors together may supply no more than half of the equity injection ³¹. Ownership is polarized: companies with ten or fewer stores own 63.0% of U.S. stores and no band between 11 and 500 stores holds even 6% ¹, so the likely buyer of an independent store is another small operator or a large chain. Chains are buying in 2026, with Casey's purchase of 24 Pak-A-Sak stores and two other deals of 20 and 21 stores ²³, but the median 8.2 times multiple for wholesale fuel distributors ²³ prices supply businesses, not single stores, so resale value rests on the appraisal. In small towns USDA's guarantee is the alternative; its fiscal 2026 availability page showed $2,093,151,202 of funding, $1,014,367,209 obligated and $1,078,783,993 remaining ⁷⁴. Our note on underserved gas station markets covers how we test whether a rural town can carry another station.

Where it fails review. Independent files stall on paper the borrower has but the file does not: a margin borrowed from a public chain instead of the borrower's invoices and competitors' posted prices; a single supplier with no named alternative and unread termination terms; a historic release that surfaces after the study; seller books, with cash sales and owner draws, that do not reach the historical coverage floor; and an inside contribution assumed to cover a fuel shortfall without category sales, margins and a payroll schedule behind it.

How it earns. A foodservice-led store draws customers with prepared food and dispensed beverages, through a larger building with a full kitchen, its own food brand or a QSR co-brand; fuel may or may not be on the site. Food carries a gross profit share well above its sales share, 38.9% of in-store gross profit on 28.5% of sales in 2025, up from 11.9% of sales in 2005, with prepared food at 73.9% of foodservice sales ³. Casey's publishes the fullest operator benchmark: a 58.3% margin on prepared food and dispensed beverages in the first nine months of fiscal 2026 ⁷⁷, a 42.2% inside margin on $2,676.3 million of inside gross profit for the year ⁷⁸, and prepared food same-store sales up 5.2% ⁷⁹. Its fiscal 2027 outlook expects inside same-store growth of 2% to 5% with same-store gallons between minus 1% and plus 1% ⁸⁰. That is the thesis a food-led study has to prove for one site: the kitchen carries debt service even if gallons do not grow. A single-store kitchen will not match a chain of 2,944 stores ⁵¹, and the study should say how far below it the subject sits and why.

Demand and cost. Food demand follows meal occasions, not traffic past the pumps, and one in three c-store customers visits a quick-service restaurant within 30 minutes of a c-store visit ⁷⁶. The trade area therefore includes every restaurant selling the same meal, sized by daypart. No development cost or staffing standard for the format was found. 7-Eleven's New Standard store runs about 5,000 square feet ⁶¹, and chains are buying food brands rather than building them, as RaceTrac did with Potbelly's 445 locations for $556 million in 2025 ⁴⁸, a deal value that should never be divided into a cost per store. A study should carry the kitchen line, hood and fire suppression, grease management, walk-in refrigeration, seating and added parking as named budget items from a bid, with kitchen payroll shown apart from cashiers.

How lenders underwrite it. With fuel on the site the store is a gas station on SBA's special purpose list; without fuel the classification is open, since the list does not name restaurants ³¹, and the study should carry the CDC's determination. Because the kitchen earns the highest margin, a small change in food sales moves coverage more than the same change in packaged goods, so the sensitivity table should put food sales on one axis. A QSR co-brand adds a second franchise agreement, which the lender will read for term, territory and fees.

Where it fails review. The uplift is where these files fail. 7-Eleven's evidence comes in three forms: a measured $6,300 of sales per store per day for New Standard stores against about $5,500 for existing stores in the first half of 2024, 13% higher ⁸¹; its president's earnings-call statement of about 18% above the system average, with mature stores projected at 45% higher ⁸²; and its chief executive's August 2025 claim of 45% higher sales than traditional stores ⁶¹. A study that averages them, or uses 45% as if it were measured, will be challenged; we use the measured figure and treat the rest as upside. The other failures are a food program opened at stabilized sales with no ramp, a chain's expansion plan cited as local demand, a fuel line left at peak margin because the thesis is food, and kitchen costs buried in contingency.

Each of the five gas station cases tests a different part of the method on this page. The Lockhart, Texas case puts a new store on the US 183 corridor in Caldwell County, a county seat on the southern edge of the Austin metro, and shows how a study handles a growing town where a pending entrance-corridor overlay could add a specific use permit to the fuel use; the Texas market page gives the state context. The Lake Wales, Florida case sits on the US 27 corridor in Polk County, where a large travel stop already serves the highway, and shows how a competitor census separates the diesel and car markets before sizing a store under SBA 7(a).

The Opelika, Alabama case is the acquisition: an operating branded store on the Opelika I-85 corridor, financed under the change-of-ownership rules at 1.25x on historical EBITDA, and it shows what a study can and cannot add when the test runs on the seller's history. The Cartersville, Georgia case places a hypothetical new store in the I-75 Exit 296 trade area, without interstate frontage, where the interchange is already served by diesel-led operators with national restaurant brands, and shows how a study narrows the concept to the car and commuter trade the site can win. The Versailles, Ohio case tests a new store in a Darke County village with a small trade area and a large employer, where the fuel component needs a conditional use permit and the village restricts underground petroleum storage in its drinking water protection zone.

Each case ends with the lender conditions its missing evidence produced. The truck stop and travel center cases in the travel center section apply the diesel method to two Texas interchanges.