Worked Example: SBA 504 Car Wash Feasibility Study in Montana, Two Self-Serve Bays and One In-Bay Automatic

This is a worked example, not a client engagement. It follows an invented sponsor who wants to build a small ground-up wash in Billings, Montana: two self-serve bays and one in-bay automatic bay on a lot of about two thirds of an acre, financed in two layers, a bank mortgage on top and an SBA-backed debenture behind it. Every cost, volume and price is a stated assumption unless a source is cited beside it, and no business, lender or address is a party. It is written for lenders, for sponsors and for anyone asking how a feasibility study treats a bay-format wash.

At a glance. A sponsor's pro forma that borrows the central margin of a broker's published range supports exactly the requested loan at the lender's minimum. A study that builds the operating costs line by line for this one site lands well under that minimum, and the 504 structure itself has no room at that income. The restructure at the end of the case is a bank first lien alone, with more cash from the sponsor.

• Total project cost, all in: $1,899,672
• Sponsor's loan request (bank first lien plus SBA 504 debenture): $1,270,000
• Sponsor equity as a share of all-in cost: 33.1%
• Stabilized net operating income, study base case: $94,491
• Debt service coverage at stabilization (binding minimum 1.25x): 0.87x
• Revenue needed for 1.25x coverage: $343,823
• Restructured loan that passes both tests: $860,000

What a Feasibility Consultant Tests First

A lender reading a request for a small wash does not begin with the building. The first question is whether the cash the site can make in a normal year carries the debt the sponsor proposes, with a margin the credit committee can defend. A feasibility consultant therefore starts from the debt and works backward: how much net operating income the loan needs, how many in-bay washes a day and what ticket that implies, and whether any public evidence supports either number.

This example puts that question to a concrete proposal. The sponsor's request is the bank first lien plus the gross debenture, together $1,270,000, against an all-in cost of $1,899,672 and sponsor cash of $629,672, which is 33.1 percent of the cost, a share the sponsor set below an assumed ceiling of 35 percent. The request itself is not a guess. It is the loan that the sponsor's own pro forma supports at the lender's minimum, rounded down to the nearest $10,000, and Table 11 recomputes it.

The sponsor's loan request is the one its own pro forma supports at 1.25x, so on the sponsor's numbers the proposal passes with $196 to spare. The consultant's job is to ask whether the sponsor's operating margin belongs to this site. On the study's line-item base case the answer is no, coverage is 0.87x, and the required restructure is a different loan, not a tweak.

Decision. The study would size the proposal on the coverage the site's own income supports, treat loan-to-cost and the sponsor's cash as outputs of that test, and report the sponsor's pro forma beside the base case rather than instead of it.

The Montana Market for a Bay-Format Wash

Montana is a small market in population and a large one in vehicles, and both facts matter to the format in this case. The state had an estimated 1,144,694 residents on July 1, 2025 ^[1], and it registered 2,473,269 motor vehicles in 2024, including 1,494,471 trucks ^[2]. The demand lens in Table 2 turns those counts into 2.17 registered vehicles per resident in Montana against 0.88 for the United States. No source in the file explains why the Montana ratio is that high; it is context, not a volume driver.

The International Carwash Association reports a third-party study from 2020 that estimated about 17,500 conveyor, 29,000 in-bay automatic and 16,250 self-service washes nationally, many of them without employees or classed with a gas station or convenience store ^[3]. The International Carwash Association also cites a Census Bureau estimate of 17,584 car washes with paid employees in 2020, which therefore undercounts the bay formats ^[4]. For a Billings bay site the International Carwash Association's line on habit matters most: the percentage of drivers who wash at a professional car wash most often "has increased from approximately 48% in 1994 to nearly 80% today" ^[5]. The habit is established; the open question for any single site is share, not adoption.

The growth in this industry has gone to the tunnel format. As of September 23, 2026, Matthews Real Estate Investment Services had confirmed 12,029 operating conveyor locations nationally, leaving out in-bay units at fuel stations ^[6]. In October 2026 Matthews reported that for 87 percent of those roughly 12,000 locations, a rival conveyor site sat within three miles ^[7]. Matthews also tracks a slowdown and a consolidation: net additions of car wash establishments dropped to 249 in 2025, down from 991 in 2021, while the 15 biggest operators now control about 30 percent of the sites it has confirmed ^[8]. A small bay-format site is a different product at a different cost to build, and a lender should not underwrite it with tunnel economics.

Matthews put the development cost of an express wash at $3 million to $5 million in September 2023 and noted the cost was rising ^[9]. This project costs $1,899,672 all in, which is why a sponsor with limited capital would look at bays.

Decision. The study would treat Montana's vehicle ratio and the national format counts as context only, would not infer Billings volumes from them, and would require the sponsor to support the in-bay volume with a traffic count or a comparable site's records before closing.

Supply and Public Comps

No sourced count of car washes in Billings or Yellowstone County exists in this file, and the case does not invent one. What the public record supports is a national structure and one local lens, shown in the table below. Each row carries its source; none of these rows is a competitor of the project and none is named as a party.

The demand lens in Table 2 applies the national ratio of residents to bay-format washes to Yellowstone County. The 2020 study's in-bay and self-service counts together make 45,250 washes, or one for every 7,326 residents, and at that ratio Yellowstone County's 172,692 residents equal about 23.6 bay-format washes. The figure is a lens, not a count: a reason to ask the sponsor for a real inventory of competing sites within a drive-time radius.

DRB, a car wash technology supplier, states that in-bay automatic washes are typically only 40 to 60 feet long, which suits small lots ^[10]. DRB also states that most in-bay automatics operate without staff, which lowers labor cost and lets them earn revenue around the clock, every day of the year ^[11], and that self-service washes do well in both city and rural areas, where owners prefer a hands-on approach or a lower price ^[12]. A supplier sells equipment, so these statements explain the format's logic and nothing more.

Decision. The study would not assume a competitive position for this site. It would hold volume at the sponsor's judgement figure, flag it as unsupported, and let the sensitivity and break-even tables show the lender how much it matters.

Demand for Washes in Billings and Yellowstone County

The Census Bureau's county estimates count 172,692 people in Yellowstone County for July 1, 2025, up from 164,726 at the April 2020 base ^[13]. For the wider Billings metropolitan statistical area the same series gives 193,603, up from 184,162 ^[14]. Table 2 prints the growth as 4.8 percent for the county and 5.1 percent for the metropolitan area. That is steady growth in the population the wash draws on, not a forecast of wash visits.

The base case assumes 45 in-bay washes a day on an annual average, an average realized revenue of $13.00 per in-bay wash and $28,000 per self-serve bay a year. None of those three has a public anchor, and the trade magazines that publish in-bay surveys sit behind a stop the research respected. At an assumed 45 washes a day the in-bay unit serves about 0.10 washes per county resident per year, a small fraction of the county and the easier claim to defend.

Decision. The study would carry the volume and ticket as assumptions, label each as unsupported by a public source, and require a signed site-specific volume estimate from the sponsor, with its method, before the loan is sized on them.

The Site and the Concept Screen

The site is a ground-up lot of about 0.65 acre on a commercial arterial in Billings. The building has an assumed 2,640 square feet: two self-serve bays of about 16 by 28 feet, one in-bay automatic bay of 24 by 56 feet (inside the 40 to 60 feet DRB gives for the format ^[10]) and a 400 square foot equipment room. Water heating, softening, spot-free rinse, reclaim and a slab-heat boiler are in the equipment cost because the site's own methodology page names traffic access and water reclamation as a car wash's binding site constraints ^[15].

The first design had four self-serve bays, an oversized bay for pickups with trailers, recreational vehicles and boats, and an in-bay automatic, on a full acre. The screen's test is set in advance: the sponsor would not bring a concept that needs more than an assumed 35 percent of the all-in cost in cash. Table 3 runs four concepts through that test with the same market and unit inputs, changing only the format, the site and the staffing. The first three needed 41.3, 39.0 and 37.3 percent equity and were dropped. The fourth, the one this case uses, needs 33.1 percent and was the first that fit. For the record, an earlier version of this example set the in-bay ticket, the bay revenues, the wage, utilities, land and site work after a rough hand estimate of coverage; those inputs are now reset to the values written down before that estimate, and every table here uses them.

The chosen concept lets one attendant and technician run the site, assumed at 1.0 full-time equivalent for cleaning, checks and snow. Matthews and DRB both say in-bay sites rarely carry labor ^[16], ^[11], but a lender normalizes for the person who actually does the work, whether the sponsor pays a wage or does the job himself. The base case therefore pays that wage and the pro forma comparison shows what the margin would be without it.

Decision. The study would carry the C2 concept forward as the sponsor's design, would show the screen so the lender sees what was dropped and why, and would not reopen the format after seeing any coverage result.

The Program Rules: SBA 504 and Why Not 7(a)

An SBA 504 loan has two lenders on one project. A bank holds a first lien on the property and a certified development company holds a second lien through an SBA-guaranteed debenture. The regulation at 13 CFR 120.801(c) describes the debenture side as a second-lien loan of up to 40 percent of project costs plus certain administrative costs ^[17], and the bank side as the balance, with a first lien ^[18]. The usual split is 10, 50 and 40 percent among the borrower, the third-party lender and the debenture ^[19].

Three limits fix the structure for a new, single-purpose building.

• The senior lender's share of total project cost may not fall under 50 percent where the borrower has been operating for two years or less, or where the asset is limited or single purpose ^[20]. This sponsor meets both descriptions, so the model puts the bank at exactly that floor.
• The regulation measures the borrower's contribution on project cost without administrative costs. It asks 15 percent of a borrower that has operated two years or less, 15 percent where the project builds a limited or single purpose structure, and 20 percent where both are true ^[21]. A new wash in a purpose-built bay building is both, so the model assumes the 20 percent. The site's own SBA program page counts car washes among special-purpose properties ^[22], which is the basis for that assumed treatment. The development company would confirm it, and the lower 15 percent would replace the assumed 20 percent if the building were not special-purpose.
• A 504 loan is generally capped at 40 percent of total project cost plus 100 percent of eligible administrative costs ^[23], and at an outstanding balance of $5,000,000 per borrower and affiliates for a project of this kind ^[24]. The debenture in Table 7 is far below both limits: the ceiling is $716,267 and the gross debenture is $400,036.

The occupancy rule is simple here. The operating company uses the whole building, above the 60 percent minimum that applies to new construction, with permanent leasing of up to 20 percent allowed ^[25]. A feasibility study may also be a loan condition under 13 CFR 120.160(b) ^[26].

Eligible costs. A 504 project may carry a construction contingency of no more than 10 percent of construction cost ^[27], and the model uses an assumed 7.0 percent. The construction loan can be refinanced too: its repayment, with points, fees and interest, is an eligible project cost ^[28], so the model carries an assumed 3.0 percent for construction-period interest. Long-term equipment with at least 10 years of remaining life is an eligible use ^[29], which covers the in-bay package and self-serve equipment. SBA also rules out working capital and inventory as uses of 504 money ^[30], which leaves the opening reserve, the short-life equipment and the closing costs to be paid in cash.

Fees. SBA's charges under 13 CFR 120.971 are a guarantee fee of 0.5 percent on the debenture, an annual fee capped at 0.9375 percent of unpaid principal, and an optional funding fee that may reach 0.25 percent ^[31]. The model assumes the development company's processing fee at its ceiling, 1.5 percent of net debenture proceeds ^[32], and the full closing fee the debenture may finance, $10,000 ^[33]. These are administrative costs, not project costs, and may be paid from the debenture proceeds ^[34]. The development company's monthly servicing fee has a floor of 0.625 percent and a ceiling of two percent a year ^[35]; the model reads the posted rate as all in.

Rates. On SBA's description, the 504 rate follows market yields on 10-year Treasury issues plus an increment ^[36], and maturities come in 10, 20 and 25 years ^[37]. The 10-year Treasury constant maturity yield was 4.86 percent on November 6, 2026 ^[38]. After the debenture pricing that completed on November 5, 2026, SomerCor 504, Inc., a certified development company, listed its 25 year rate at 6.57 percent ^[39], which the model reads as 1.71 points over the 10-year yield. SomerCor 504, Inc. also states that rural-area businesses pay about 25 basis points less because two SBA fees are waived in fiscal 2027 ^[40]. The case does not apply that waiver: a Billings metropolitan site is modeled at full fees and 6.57 percent, and SBA's fee notice never reached this file ^[41], so the case cannot say how SBA defines rural.

For the bank, the base is the 5-year Treasury yield that closed at 4.64 percent on November 6, 2026 ^[42]. Adding an assumed margin of 2.75 percent gives the 7.39 percent the model prints, amortized over an assumed 25 years.

Jobs. Under SBA's September 30, 2025 notice each $95,000 guaranteed must create or retain one job opportunity ^[43], a rule that covers 504 loans approved from October 1, 2025 ^[44]. At that figure the debenture calls for 4.2 job opportunities and the wash employs 1.0 full-time equivalent. With a single person on the payroll the wash cannot reach that count, so its eligibility would rest on the route for projects meeting community development or public policy goals, open if the development company's 504 loans as a whole, this one counted, meet or exceed the company's required job average ^[45]. The lender should ask the development company for that portfolio average early.

The coverage floor. For a 504 loan this example reads its program floor from a single line of the site's own SBA program page: "504 loans need 1.15x on history or two-year projections" ^[46]. No SOP text stands behind that figure here; it is the page's statement and nothing more. Against it sits the lender's own assumed 1.25x minimum. The lender's figure is the higher, so for this Billings wash the binding minimum is the lender's assumed 1.25x.

Why not 7(a). Suppose the assumed request of $1,270,000 were a 7(a) loan. Prime stood at 7.00 percent from November 2 through November 6, 2026 ^[47], and the rate would sit above it by an assumed 2.75 points, under the 3.0 point ceiling for 7(a) loans above $350,000 ^[48], giving 9.75 percent. The model assumes the maximum guaranty for a 7(a) loan above $150,000, 75 percent ^[49], and the maximum term, 25 years ^[50]. The guaranty fee at the statutory ceiling of 3.5 percent of the guaranteed share for a loan above $700,000 ^[51], plus 0.25 percent of the guaranteed share above $1,000,000 ^[52], comes to $33,338. Table 4 shows the result: annual debt service of $135,809 and coverage of 0.70x, against $109,025 and 0.87x under the 504. The 7(a) loan's one structural advantage is that it could fund the reserve and short-life equipment, which a 504 cannot.

All inputs, assumed and sourced, are listed in Table 5.

Rounded to the dollar; lines may not foot by $1.

Decision. The study would structure the proposal as a 504 because the lower fixed-rate debenture carries a lower payment than 7(a) on the same loan amount, and would compute the supported loan against both the lender minimum and the program floor.

SBA Lending Record for This Asset Class in Montana

Loan Analytics keeps a loan-outcomes dataset built from the SBA's own loan-level 7(a) and 504 files ^[70]. For this example's state and asset class it covers SBA loans approved in fiscal 2017 to 2026 to businesses coded to car wash, with the project in Montana. The figures are the SBA program's record in the state, not Loan Analytics' work and not this hypothetical project. The dataset notes a limit on the class: car and truck washes, including mobile washes, share one code.

What the Wash Costs and Who Pays: Sources and Uses

Table 7 builds the project from the ground up. Land is an assumed $10.90 per square foot on 0.65 acre, $308,623, with no Yellowstone County land sale in the file to support it. Site work is an assumed $9.10 per square foot of lot, $257,657, covering paving, curb cuts, water and sewer service, stormwater, lighting and landscaping. The bay building is 2,640 square feet at an assumed $215.00 per square foot, $567,600. That price sits below the range Rider Levett Bucknall publishes for Denver, the nearest market on its list, which runs $235 to $350 per square foot for strip retail ^[55] in indicative hard costs, measured by gross floor area ^[71]. The reasoning for going lower is that a bay building has no tenant finish, though it carries heated slabs, trench drains and wash-grade masonry; a lender could fairly ask for a bid. Rider Levett Bucknall also reports that its Denver index rose 4.55 percent from July 2025 to July 2026 ^[72].

Wash equipment, water treatment and reclaim are an assumed $434,000, built from a $290,000 in-bay package, $42,000 per self-serve bay and $60,000 for water heating, softening, reclaim and the boiler. Architecture, engineering, permits and utility connections are an assumed 9.0 percent of site work and building, $74,273. The assumed contingency is $57,768, or 7.0 percent of construction cost against the 10 percent cap, and construction interest is $40,006. These make a 504 project cost of $1,739,927.

How the money is split. Under the 50 percent minimum, the bank first lien is $869,964. The debenture is $379,740 net to the project, 21.8 percent of the 504 project cost, and the borrower's contribution to the project is $490,224, or 28.2 percent against the 20 percent minimum. Administrative costs of $20,297 are financed in the debenture, which makes it $400,036 gross: the SBA guarantee fee of $2,000, the development company's processing fee of $5,696, the funding fee of $1,000, the closing fee of $10,000 and an underwriters' fee of $1,600, the last at an assumed 0.40 percent because the regulation sets no rate for it.

Cash outside the 504. The borrower also pays $49,000 for short-life equipment, an assumed $60,000 opening reserve and $30,449 in third-party reports and closing costs, all cash outside the 504 project, together $139,449. Added to the 504 borrower contribution this puts sponsor cash at $629,672, and the sources match the all-in cost of $1,899,672.

The reserve is a weak spot. Table 7 shows it falls $31,459 short of the shortfall after debt service that the ramp in Table 10 produces over the first two years. A lender would want the sponsor to fund a larger reserve or to show a liquid outside source, since a 504 cannot pay for working capital.

Decision. The study would carry the all-in cost of $1,899,672 as the cost to be financed, would ask for a contractor's bid and the city's utility letters to replace the judgement lines for land, site work and building, and would size the opening reserve to the ramp shortfall, not to a per-bay rule of thumb.

The Pro Forma: Revenue, Operating Costs and the Sponsor's Version

Revenue. Table 8 builds stabilized revenue of $278,525. The in-bay automatic washes 45 cars a day, or 16,425 a year at an assumed $13.00, for $213,525, which is 76.7 percent of the total. The two self-serve bays bring an assumed $56,000, and vacuums, vending and other sales bring an assumed $9,000. The in-bay unit carries the site, and that concentration is why the break-even table speaks mainly in in-bay washes a day.

The sponsor's version. A sponsor would typically start from a published margin. Car Wash Advisory, a car wash M&A advisory firm, states that in-bay automatic and self-service car washes can generate 50 to 67 percent EBITDA (adjusted cash flow) margins for the owner ^[54], and that a single-site owner of such a wash can expect 50 to 60 percent ^[53]. It uses adjusted cash-flow margin and EBITDA margin as the same measure and puts car washes generally at 40 to 65 percent depending on format, volume and the number of sites owned ^[73]. The model assumes the central value of the single-site range, 55.0 percent, for the sponsor's pro forma only. Applied to revenue of $278,525, and after the assumed 3.0 percent management charge and assumed 3.0 percent reserve that both columns deduct, the sponsor's net operating income is $136,477. Against the combined payment of $109,025 that is 1.25x, which is how the assumed request of $1,270,000 was found.

The study's version. The study does not take a margin from a broker's range. It builds each operating cost for this lot, as Table 9 prints. Labor, with payroll taxes and benefits, is $41,860, which is 15.0 percent of revenue, at the low end of the 15 to 20 percent ceiling that, according to Matthews, the most valuable washes keep to ^[66]. Chemicals are an assumed $19,497. Utilities, at an assumed 12.0 percent of revenue, are $33,423. Repairs are $16,712, snow removal and grounds $5,850, card processing and software $8,356, marketing $5,571, real property tax $11,055, assumed insurance of $15,000 and assumed accounting, licenses and alarm of $10,000. Operating expenses before management and reserve are $167,323, an EBITDA margin of 39.9 percent. After management of $8,356 and the replacement reserve of $8,356 the net operating income is $94,491, a margin of 33.9 percent.

Why the two differ. The study's margin sits 15.1 points below the assumed central value of Car Wash Advisory's single-site range. Revenue, the cost basis and the rates are the same in both columns; only the operating cost build differs. Four items explain most of the gap.

1. Paid site labor. An owner-operator who washes, repairs and cleans the site himself shows the broker's margin, because his pay is the profit. A lender normalizes for the work, and with the wage added back the study's EBITDA margin rounds to 55.0 percent, level with the assumed central value. So the labor line accounts for almost the whole difference.
2. Billings winter. Slab and space heat sit inside the assumed utilities line of 12.0 percent of revenue, and snow removal is an assumed $9,000 per acre. No Montana source fixes either number; they are judgements for a cold climate, and the case does not pretend otherwise.
3. Fixed cost on a small revenue base. Car Wash Advisory puts insurance at roughly $15,000 per site ^[65], which the model assumes; on revenue of $278,525 it is a heavy fixed charge, where a larger site would spread it thinner. The property tax is on a new building valued at cost.
4. Chemistry. Car Wash Advisory states that chemicals on average represent 8.5 to 10 percent of a car wash's total costs ^[58]. The model's assumed 7.0 percent of revenue makes them 10.6 percent of all operating costs, a little above that range, because an in-bay machine uses more chemistry than a wand.

Montana's rules, in the cost lines. Wage is an assumed $17.50 an hour plus a 15 percent burden, above the state minimum. The Department of Labor and Industry gave notice on October 7, 2026 that the minimum rises to $11.20 an hour on January 1, 2027 ^[57]; the current minimum is $10.85 ^[74], adjusted yearly for CPI-U inflation ^[75].

Property tax follows the state's new tiers. The Revenue Interim Committee staff paper lists, for 2026 and after, a class four commercial and industrial rate of 1.50 percent below six times the median value and 1.90 percent at or above it ^[61], with tiered rates based on the statewide median commercial property value after Senate Bill 542 and House Bill 231 ^[60]. The model applies the 1.50 percent to the whole value, since the median was not sourced, with an assumed levy of 650 mills that no local source fixes, so the $11,055 line is a judgement. Wash equipment of a service business is class eight property ^[62], ^[63], and the first $1 million of its market value is exempt ^[64], so the model treats the $483,000 of equipment as exempt.

The ramp. Table 10 shows stabilization taking three years. Year 1 carries 75 percent of stabilized revenue and Year 2 carries 90 percent, both assumed. Coverage is 0.46x in Year 1, 0.70x in Year 2 and 0.87x in Year 3, with a cash shortfall after debt service of $59,099, $32,360 and $14,535. The site does not cover its debt service in any of the three years. On the site's own SBA program page the 504 floor is stated as 1.15x on history or on two-year projections ^[46], and Year 2 coverage of 0.70x does not reach it.

Decision. The study would underwrite the operating statement built from line items and would show the broker-margin pro forma beside it, because a lender needs both: the first tests the site, the second tests the sponsor's claim.

Coverage Under the Convention, Sensitivity and Break-Even

Table 11 shows the combined annual debt service of $109,025: $76,402 on the bank first lien at 7.39 percent over 25 years and $32,623 on the debenture at 6.57 percent over 25 years. The loan constant is 8.58 percent. The sponsor's column shows coverage of 1.25x, equal to the assumed lender minimum, and the study's column 0.87x.

The convention for these cases tests coverage against two minimums. The first is the program floor, which the site's own SBA program page states as 1.15x for 504 loans ^[46]; the page is cited and nothing in the SOP text was consulted. The second is the lender's minimum, an assumed 1.25x. Of the floor and the lender minimum, the larger sets the binding figure, and the assumed lender minimum binds here, at 1.25x. The sponsor's column clears it by $196 of net operating income. The study's column misses it by $41,791, falls below the 1.15x floor, and falls below 1.00x, which means the site does not cover its debt at all.

The supported loan, with its equity. The loan that the study's income supports at the binding minimum, in the same mix, is $880,553. On that smaller loan the recomputed debenture fees are $17,094, the project would cost $1,896,470 in total, and the sponsor's equity would be $1,015,916, or 53.6 percent. They also expose a second problem. At $880,553, with the same mix, the bank first lien shrinks to 34.7 percent of the 504 project, short of the 50 percent a new business or single-purpose asset requires ^[20], so the same-mix loan is not a valid 504.

Sensitivity. Table 12 flexes one input at a time against the study's base case. Coverage reads 0.70x with revenue down 10 percent, 0.54x with revenue down 20 percent, 0.73x with rates up 2 points and 0.70x with operating expenses up 10 percent. The base case already fails, so the table's use is the other direction: revenue is the strongest lever, and the bank rate matters more than it looks because the bank first lien carries the larger payment.

Break-even. Table 13 turns the miss into a size. Variable costs run at 36.0 percent of revenue, fixed operating costs are $83,765, and the annual debt service is $109,025. Revenue needs to reach $301,235 for 1.00x, which is 108.2 percent of the base case; $326,788 for the 1.15x floor, 117.3 percent; and $343,823 for the binding 1.25x, 123.4 percent. In in-bay washes a day, with self-serve and other revenue at base, that is 50, 55 and 59 against 45 assumed. Alternatively, holding the wash count at 45, the average in-bay revenue per wash would need to be $16.98 against the assumed $13.00.

Decision. The study would report the base case at 0.87x as a miss of the binding minimum, would not move any input to close it, and would hand the lender the break-even volumes and the restructure instead.

What Would Change the Answer

Three inputs decide this file, and they are not equally in the sponsor's control.

First, in-bay volume and ticket. Neither the assumed 45 washes a day nor the assumed $13.00 ticket has a public anchor; the break-even table shows that 59 washes a day, or $16.98 a wash at the assumed count, reaches the binding minimum. The sponsor can test those numbers with a local traffic count, a competitor inventory and a pricing plan.

Second, the gap between the study's operating costs and the broker's margin. Most of it is paid labor and winter cost. If the sponsor will run the wash himself and the lender accepts an owner-operator's labor as a distribution, the labor add-back brings the EBITDA margin to 55.0 percent, but then the coverage test rests on a person who cannot also run another business and cannot be replaced at that price. Lenders often decline that add-back, and the base case does not use it. A cost bid for heat and snow removal from a local contractor would replace two judgement lines.

Third, the bank's rate, since the bank first lien carries the larger of the two payments. The rate is the 5-year Treasury plus an assumed 2.75 points. A 2-point rise takes coverage to 0.73x, which is why a lender would want the fixing date and the debenture sale date written into the closing schedule. That Treasury base stood at 4.64 percent on November 6, 2026 ^[42] and will not wait for the closing.

There are other risks a lender would list: permitting and construction time of 18 months or more, in Matthews's March 2026 words ^[76], buyers' aversion to new builds, also reported by Matthews, which affects the collateral's liquidity ^[67], and a special-purpose building that the site's own SBA program page treats as having limited alternative use ^[22].

The restructure. The 504 does not help at this income, and the reason is arithmetic. At $94,491 of net operating income, the debt service the income carries at the assumed binding minimum of 1.25x is $75,593. The bank first lien at the 504's 50 percent minimum share would cost $76,402 a year on its own, more than the income carries, so the debenture the remaining capacity would carry is $0, and a debenture of any size would also carry the closing fee, which the model assumes at its $10,000 maximum. The bank's required share uses up all of the coverage the income can bear before the second lien starts. The restructure therefore drops the debenture. It is a bank first lien alone, at the same 7.39 percent over 25 years, sized to the assumed 1.25x minimum and rounded down to $10,000: $860,000, which is 49.4 percent of the project cost.

The restructure passes both tests. Annual debt service on $860,000 is $75,527, so coverage is 1.25x, at the assumed binding minimum of 1.25x and above the 1.15x floor that the site's SBA program page gives for 504 loans ^[46]. The restructured loan has no program behind it, so the lender minimum governs; the floor is printed for comparison. With no 504 administrative costs the total project cost is $1,879,376, and the sponsor's equity is $1,019,376, or 54.2 percent of the cost, which is $389,703 more than at the request. This is a computed restructure, not a concept the sponsor would bring under the assumed 35 percent cap: it asks for 54.2 percent where the sponsor was willing to put in 33.1 percent. Either the study's costs would have to come in near the comparable's margin, or revenue would have to reach 123.4 percent of the base case, as the break-even table shows.

Decision. The study would report the 504 as not viable at this income, would present the bank first lien alone as the only structure shown to pass both tests, and would tell the sponsor that the next dollar of effort belongs in the volume plan, not in the loan structure.

Frequently Asked Questions About Choosing a Feasibility Study Company

What does the SBA's own record show for a car wash loan in Montana? The SBA's loan record for this class in the state is shown in Table 6. It counts 23 loans in fiscal 2017 to 2026, with a median of $445,000. The request here, $1,270,000, is larger than the median and sits inside the middle half of the loans, which runs from $128,300 to $2,457,000. The record is the program's history in the state, not a forecast for this project, and 504 approvals alone are too few to publish as their own file.

How would I choose a feasibility study company or consultant for a car wash? Ask whether the consultant builds the operating statement from line items for your site or applies a margin from a broker's range. The difference between those two methods is the whole gap in this example. Ask for the break-even in washes per day, since that is the number you can test with a traffic count. Ask whether the work states each assumption and each public source separately, whether it reports a coverage miss as a miss, and whether it names the gaps in the evidence, such as an unsourced local land price or levy, instead of filling them silently.

Why would the study use a lender's 1.25x when the program floor is 1.15x? The lender's minimum is an assumption here, not a program rule. The site's own SBA program page states 1.15x for 504 loans ^[46]; the SOP text was not part of the reading for this case.

Why is the sponsor's pro forma different from the study's base case? They share revenue, cost basis and rates and differ only in the operating costs. The sponsor's pro forma applies an assumed 55.0 percent, the central value of the Car Wash Advisory single-site range ^[53], and gets $136,477 of net operating income. The study builds line items for the lot and gets $94,491. The paid labor, winter costs and fixed costs on a small revenue base explain the difference, as the pro forma section shows.