Best Franchises to Own in 2026: What the SBA Loan Data Actually Says

Every winter the franchise rankings arrive on schedule. Entrepreneur scores its Franchise 500 from franchisor questionnaires. Franchise Times adds up systemwide sales. Satisfaction surveys get tabulated, growth stories get polished, and several hundred brands are told they are among the best investments in America. Almost none of it rests on the one dataset that records what actually happened to the people who borrowed money to buy in.

That dataset exists, and it is free. The U.S. Small Business Administration publishes a loan-level file for every 7(a) and 504 loan it has guaranteed since 1991, refreshed quarterly, with the franchise brand recorded on the file (2). Roughly one in five SBA loans goes to a franchisee, and by FRANdata's estimate more than $15 million in SBA credit flows to franchise businesses every single day (3). When those loans get repaid, the record says so. When they get charged off, the record says that too. It is the closest thing American franchising has to a report card graded by someone other than the school. And the scale is not small: since fiscal 2010 alone, the file carries more than 92,000 7(a) loans flagged to a franchise brand, $65.5 billion in approvals, with another 11,390 franchise 504 loans worth $14.6 billion behind them (2).

The grades complicate the sales pitch. For this article we ran the computation ourselves on the March 31, 2026 release of the file: among 7(a) loans approved from fiscal 2010 through fiscal 2019 that have since resolved, franchise-flagged borrowers charged off at 9.4 percent, 3,337 failures out of 35,406 finished loans, against 7.1 percent for everyone else (2). An independent tracker running a later release of the same file lands at 10.2 versus 7.8 percent (1). Take either vintage; the direction is identical. On average, buying a brand did not buy safety.

Averages are where the story starts, not where it ends. Inside that 9.4 percent sit brands that charged off two loans out of every three and brands that have barely charged off any. Categories diverge even more than brands do. And 2026 is an unusually consequential year to be reading this data, because the rules of SBA franchise lending have changed more in the past fifteen months than in the previous decade: the Franchise Directory came back, a certification deadline just culled the eligible list, equity requirements tightened, and a new SOP arrives on October 1 carrying a debt-coverage test that will reshape acquisition lending (4)(5)(6).

So this is a different kind of best-franchises list. It ranks the 2026 opportunity set the way a lender reads a file: loan performance first, then unit economics, then momentum, with the marketing held at arm's length.

How to read loan data without fooling yourself

A charge-off is not a bad quarter. It is the end of the line: the borrower stopped paying, the collateral was liquidated, and the SBA wrote a check to the lender on its guaranty. So a brand-level charge-off rate is a hard-outcome statistic, which is exactly what makes it valuable and exactly why it gets misused.

Four rules keep the reading honest. First, only resolved loans belong in the denominator. A rate computed as charged-off loans divided by loans that have either been paid in full or charged off answers a real question: of the borrowers whose story is finished, how many failed? Mixing in active loans flatters everyone. Second, cohorts need seasoning. A loan approved in 2023 has barely had time to fail, which is why the cleanest current benchmark uses the fiscal 2010 through 2019 approval cohort (1). Third, small denominators produce noise, not insight. A brand with nine resolved loans and three charge-offs has a 33 percent rate and no statistical meaning; serious readings start at a few dozen resolved loans per brand. Fourth, and least appreciated, loans are tagged to a brand at origination. If a system was sold, rebranded, or turned around in 2019, its 2012 failures still sit in the file under the old flag. That matters enormously for the caution list later in this piece.

One more calibration. The Federal Trade Commission, working a different window and denominator on the same underlying records, found franchise borrowers defaulting at 3.9 percent against 3.5 percent for non-franchise borrowers, and cautioned that SBA borrowers are a credit-screened subset that may not represent franchise buyers financed outside the program (7). The absolute level moves with the methodology. The direction does not: across every serious cut of the federal data, franchisees have not outperformed independents, and in the seasoned cohorts they have underperformed them (1)(7)(8). The SBA itself publishes no brand-level default table, a gap two Senate bills have tried to close by mandating quarterly brand-level disclosure (9). Until that happens, the loan tape is the source, and reading it correctly is the skill.

A note on the numbers in this article that carry no outside attribution: they are our own calculations from the SBA file (2), March 31, 2026 release, computed exactly as described above. Resolved means paid in full or charged off; loans currently sitting in purchase or liquidation status are excluded from both sides, which makes our rates conservative. Brand names were normalized and obvious variants merged before counting, since the file spells Subway at least two ways and Dunkin' three. Thresholds are 100 resolved loans for the best-record table and 40 for the caution list. And one structural caveat travels with everything: only 8.7 percent of FY2010-2019 records carry a franchise name, against a true franchise share of the program closer to one loan in five (3), so these rates describe the loans lenders flagged, not every franchise loan ever made.

This measures loan outcomes. It does not measure franchisee happiness, unit profitability for owners who never borrowed, or what a brand will do next. Keep that boundary in mind and the data is the most useful thing in franchising. Ignore it and the data becomes a blunt weapon.

The 2026 rulebook changed. Read it before the rankings.

The regulatory ground under franchise lending shifted three times in three years, and anyone shopping brands in 2026 needs the current map, not the 2023 one still circulating in broker decks.

Start with the SBA Franchise Directory. The agency eliminated it in May 2023 and pushed franchise eligibility decisions onto lenders. Effective June 1, 2025, under SOP 50 10 8, the Directory came back with teeth: if a brand meets the Federal Trade Commission's definition of a franchise, it must be listed on the Directory for its franchisees to obtain SBA financing at all (4)(5). No listing, no loan. That single sentence restored the SBA as gatekeeper and made a thirty-second Directory lookup the first diligence step on any brand purchase.

Then came the cliff. Franchisors already on the Directory had to execute a new SBA Franchisor Certification, a deadline extended twice and finally fixed at June 30, 2026; brands that failed to certify faced removal, and removal ends 7(a) eligibility for their buyers (6). A brand can have healthy stores, a glossy development team, and no path to SBA financing because someone in its legal department missed a filing. In 2026, Directory status is not paperwork trivia. It is a gating condition on your capital stack.

The affiliation rules moved in the opposite direction, toward simplicity. Under the current 13 CFR 121.301, affiliation for the loan programs is determined by ownership, full stop (10). The old control analysis, in which territory clauses, transfer restrictions, and franchisor purchase options could sink a deal, is gone from the loan programs, and the SOP states outright that franchise agreements no longer need to be reviewed for control (4)(10). Plenty of online guides still recite the old test. They are wrong, and following them wastes negotiating capital on provisions the SBA stopped caring about.

Money rules tightened. Startups and complete changes of ownership now require a minimum 10 percent equity injection against total project cost, and a seller note only counts toward it if it sits on full standby, no principal and no interest, for the life of the loan (4)(11). The FICO SBSS prescreen for smaller 7(a) loans was retired effective March 1, 2026, pushing lenders back to full manual underwriting on files they used to score in minutes (12).

And one more turn is coming. On August 14, 2026 the SBA issued SOP 50 10 8.1, effective for loans numbered on or after October 1, 2026. As reported from the issuing notice, it carries the Directory regime and equity rules forward and adds two acquisition-side screens: a 1.25x historical debt-service-coverage test for first-time business acquisitions and a lender-ordered quality of earnings report on deals of $3 million and up (13). If you are buying an existing franchise unit, the target's last three tax returns just became the whole ballgame.

A market splitting in two

Franchising as a whole is still growing, just not the way the promotional decks say. The International Franchise Association's 2026 outlook, produced by FRANdata, projects establishments rising from 832,521 to roughly 845,000 units, a 1.5 percent gain, with employment approaching 8.9 million jobs and output reaching $921.4 billion (14)(15). Modest, respectable, and notably humbler than the year before. The 2025 edition had forecast 851,402 units by year-end; the actual base came in some 19,000 units short (16). When the industry's own economists overshoot by that margin, treat every forward projection in this business, including these, as directional rather than gospel.

What the aggregate hides is a hard split by category. FRANdata's 2026 projections put child services and commercial and residential services, the home-services complex of HVAC, plumbing, roofing, restoration, and cleaning, in a tie for fastest growth at 3.2 percent each. Retail food follows at 2.3 percent, health and wellness at 2.1 percent, full-service restaurants at 2.0 percent, personal services and lodging at 1.8 percent, business services at 1.6 percent. Quick service restaurants, automotive, and real estate sit below half a percent (14)(15).

Meanwhile the money is telling its own story. With prime at 6.75 percent, variable 7(a) pricing runs roughly 9.75 to 13.25 percent APR, and the overall 7(a) program contracted sharply in fiscal 2026, down by a third in loan count through nine months. Franchise lending moved the other way: 7(a) lending to the top-100 franchise cohort rose 7.4 percent by count and 51.3 percent by dollars in the first half of the fiscal year, even as non-franchise lending fell (20). Lenders squeezed by a tougher SOP are concentrating in standardized, well-documented concepts. That flight to legibility is the quiet tailwind behind every ranking on this page: the right brand is not just a business decision in 2026, it is an approval-odds decision.

What it costs to get in, and why loan sizes differ

Before naming names, one mechanical truth explains most of what follows: the franchise fee is a rounding error, and the building is the investment. Across an analysis of 2,185 current franchise disclosure documents, the median initial fee is $40,000, barely a tenth of the median total project midpoint of $362,381 (21). What separates a $150,000 loan from a $3 million loan is real estate and buildout, not the brand's sticker price.

The category medians make the ladder visible. Food and beverage concepts run a median $358,500 to $822,250 in Item 7 initial investment. Fitness runs $309,249 to $710,900. Pet services $194,750 to $470,450, automotive $188,350 to $622,500. Then the floor drops: home services $128,368 to $227,409, child services and education $124,025 to $316,824, senior care $118,030 to $242,840, business services $77,500 to $152,100 (21). A drive-thru restaurant borrower and a home-services borrower are not making different versions of the same bet. They are making different bets, at different leverage, with different failure costs.

The brand tables: what the loan file itself says

Everything in this section is our own computation from the loan-level file, March 31, 2026 release, seasoned FY2010-2019 approval cohort, resolved loans only (2). No franchisor supplied a number, and every figure shows its denominator.

Start with the honor roll. Among brands with at least 100 resolved cohort loans, these posted the cleanest records:

Brand	Charge-offs / resolved	Rate	Median loan
Culver's	0 / 176	0.0%	$715,000
Club Pilates	0 / 148	0.0%	$191,800
Primrose Schools	0 / 142	0.0%	$3,090,200
Christian Brothers Automotive	0 / 137	0.0%	$309,000
Nothing Bundt Cakes	0 / 115	0.0%	$350,000
Planet Fitness	0 / 111	0.0%	$1,010,000
The UPS Store	2 / 415	0.5%	$177,100
Wingstop	1 / 146	0.7%	$350,000
Domino's Pizza	2 / 185	1.1%	$250,000
Dunkin'	3 / 257	1.2%	$472,000
Orangetheory Fitness	5 / 308	1.6%	$463,500
Great Clips	4 / 221	1.8%	$149,200
Home Instead Senior Care	3 / 170	1.8%	$295,000
Jersey Mike's	5 / 223	2.2%	$300,000
Kiddie Academy	4 / 156	2.6%	$671,000
The Goddard School	6 / 219	2.7%	$868,900

Two patterns jump out. Education and child care brands cluster near the top at school-sized loans, three of them with six- and seven-figure medians and next to no failures. And the hotel flags, which we left off the table because franchise naming is messiest there, tell the same story at the largest loan sizes in the file: Days Inn at 1.6 percent, Quality Inn at 1.8 percent, Comfort brands at 1.3 percent, all on median loans between $1.7 million and $2.7 million (2). Big loans are not the risk. Big loans against thin volumes are.

Momentum is the second table. Counting 7(a) approvals from fiscal 2023 through fiscal 2025, franchise-flagged lending ran 7,363, then 5,904, then 7,135 loans a year, and the brands lenders financed most often were these (2):

Brand	Loans FY2023-2025	Median loan
The UPS Store	514	$345,000
HOTWORX	243	$439,000
Naturals2Go	161	$150,000
Scooter's Coffee	149	$1,215,000
GameDay Men's Health	145	$246,600
Subway	142	$222,500
Anytime Fitness	128	$373,950
Crumbl	119	$575,100
Quality Inn (Choice Hotels)	118	$2,936,700
Domino's	112	$629,000
Tropical Smoothie Cafe	110	$500,000
The Goddard School	107	$1,325,000
Biggby Coffee	103	$400,000
The Learning Experience	100	$496,250

Read the two tables together and the loan-size mechanics from the capital ladder come alive: Scooter's median financed deal is $1.2 million because the kiosk is the collateral, Goddard's is $1.3 million because the school is, and The UPS Store tops the volume chart at $345,000 because an inline print-and-ship counter is exactly the kind of legible, modest credit a 2026 lender wants to write. Brand names in both tables are normalized from the file's raw spellings.

The category floor matters as much as any brand. In the same seasoned cohort, franchised restaurants charged off at 11.4 percent (933 of 8,213 resolved), fitness at 11.6 percent (291 of 2,516), auto repair at 11.7 percent (112 of 960), home health at 6.9 percent (41 of 593), and child care at 2.6 percent (21 of 804), the cleanest category in the dataset (2). Every brand judgment that follows sits on top of those baselines. The full brand table, every franchise with at least 20 resolved cohort loans, 372 brands in all, is downloadable at the end of this article; cite it freely with a link back.

The 2026 category read

Food and beverage: a shrinking tide with a few fast boats

The honest headline is caution: QSR sits below half a percent projected unit growth, the bankruptcy docket is busy, and this is the category where debt service meets discretionary spending head-on (14)(17). The brands worth owning here are the ones whose unit volumes make the math boring.

Culver's is the cleanest example in America. Its 2026 disclosure puts franchised average unit volume at $4,142,737 across 988 restaurants, up 9.3 percent in a year, with the system adding a net 44 units in 2025 (22). The catch is the ticket to play: Item 7 now runs $3,410,000 to $10,290,000 including land, a range that jumped roughly 29 percent at the low end in a single year on construction and land inflation (22). At 2026 borrowing costs, that is a seven-figure project underwritten against a four-million-dollar volume, which works precisely because the volume shows up. The loan file agrees: Culver's franchisees resolved 176 cohort loans at a median $715,000 without a single charge-off, the best record of any brand its size in the dataset (2). Few brands can say the same.

Drive-thru beverage is the growth pocket. 7 Brew's 2025 disclosure shows franchised stands averaging $1,989,229 in gross sales with store-level EBITDAR near 29 percent, against a build of $894,000 to $2,178,500, and the system has roughly tripled its footprint since 2023 (23). Scooter's Coffee pairs a kiosk build of $658,898 to $1,345,750 with average volumes around $880,000 and added 83 net stores last year (24). The risk in both is the same risk: a purpose-built drive-thru is a special-purpose box financed at special-purpose loan sizes, and the margin for a slow ramp is thin. Anyone modeling one of these should treat the first-year revenue curve, not the mature AUV, as the underwriting question; it is the exact problem a lender will probe inside a restaurant and QSR feasibility study.

And then there is the cautionary giant. Subway posted a record average unit volume near $490,000 in 2024, and still shrank for the ninth straight year, a net 631 U.S. closures in 2024 and another 729 in 2025 (25)(26). Its enormous historical footprint means Subway will dominate any raw count of SBA franchise defaults almost by definition, and the file makes the distinction cleanly: on 1,197 resolved cohort loans, by far the most of any brand, Subway's charge-off rate is 7.2 percent, a shade below the franchise average (2). That is the lesson in reading this data: footprint drives counts, unit economics drive rates.

Home services: the category the decade is built for

If a committee designed a franchise category for the 2026 lending environment, it would look like this: 3.2 percent projected unit growth, entry costs of $128,368 to $227,409, and demand anchored in physics rather than fashion (14)(21). FRANdata projects roughly 119,000 franchised home-services establishments producing $143.3 billion in output this year (27). Behind it sits the oldest housing stock in American history, a median home age of 44 years, with more than four in ten homes past fifty (28). Roofs, pipes, compressors, and crawl spaces do not care about consumer sentiment.

For a borrower, the shape of the deal is the appeal. Territory and van-based models mean small loans, fast breakeven, and no seven-figure buildout standing between opening day and cash flow. For a lender, the tradeoff is thin collateral, which the modest leverage largely answers. The operating risk is labor: the skilled-trades shortage that inflates ticket prices also makes technicians the scarce input, and the operators who fail here usually fail at hiring, not at demand. Servpro's cohort record, 13 charge-offs on 287 resolved loans for a 4.5 percent rate, shows the model working at scale (2), though the caution table below shows the category is not immune when a small territory concept meets a thin service line.

Child care and early education: fastest growth, real regulation

Child services ties home services for the fastest projected growth in franchising at 3.2 percent, on entry costs of $124,025 to $316,824 (14)(21). The demand driver is structural: dual-income households have turned licensed care into infrastructure, and in most metros supply has not caught up. The complication is that this is a licensed, inspected, staffing-ratio business where the regulatory file is as important as the P&L, which is why lenders scrutinize these deals the way we detail in our childcare feasibility work.

The loan file is emphatic about the category. Franchised child care resolved 804 cohort loans at a 2.6 percent charge-off rate, the cleanest category in the dataset, and the flagship school brands carry it: Primrose at zero for 142 on a $3.1 million median loan, Goddard at 2.7 percent on 219, Kiddie Academy at 2.6 percent on 156 (2). Lenders have noticed; Goddard and The Learning Experience both sit in the top fifteen most-financed franchises of the past three fiscal years (2).

The education-adjacent tutoring segment shows how legacy loan data needs date-stamping. Huntington Learning Center, entry $159,367 to $298,357 with a stout 9.5 percent royalty, appears on 2010s-era high-default lists yet operates steadily today at more than 250 centers (29). Old cohorts describe old systems. Current disclosure describes the one you would buy.

Senior care and home health: demographics versus labor

No category has a cleaner demand curve. More than 73 million Americans will be 65 or older by 2030 (30), the Bureau of Labor Statistics projects 17 percent growth for home health and personal care aides through 2034 with roughly 766,000 openings a year (31), and 77 percent of adults over 50 say they intend to age in place (32). Entry costs are the lowest of any major category, $118,030 to $242,840 (21), which means small loans and cheap failures.

The honest caveat mirrors home services, amplified: the caregiver is the product, and the same BLS numbers that prove demand also prove the hiring problem. Margins are thinner than the brochures imply, and the operators who thrive are the ones who treat recruiting as the core business. As an SBA credit, though, the profile is attractive: low leverage, contract-adjacent revenue, and a demand base that compounds for twenty years. The cohort record backs the read, with Home Instead at three charge-offs on 170 resolved loans and franchised home health as a whole at 6.9 percent, comfortably below the franchise average (2).

Fitness: own the barbell, avoid the middle

Fitness runs above the franchise average on charge-offs, 11.6 percent across 2,516 resolved franchise loans in the seasoned cohort by our count (2), and the past two years explain why: Blink Fitness took its corporate clubs through Chapter 11 in August 2024, and the onetime largest Orangetheory franchisee wound up in Chapter 7 (33). The mid-market gym, too premium to be cheap and too cheap to be premium, is where the bodies are buried.

The loan file draws that barbell in ink. Zero charge-offs on 111 resolved Planet Fitness loans, zero on 148 for Club Pilates, Orangetheory at 1.6 percent; then mid-market Anytime Fitness at 12.2 percent on 589 resolved loans and Snap Fitness at 9.9 percent; then the boutique bust-outs, iLoveKickboxing at 28.7 percent, 9Round at 24.7 percent, F45 Training at 19.7 percent, filling out the caution tail (2).

The scale end still performs on the disclosure side too. Planet Fitness reports median franchised club revenue around $1.79 million across 2,291 clubs, against a build of $1,282,500 to $5,386,000 and a 7 percent royalty (34). That is big-box economics with utility-like membership revenue, and it is effectively a multi-unit operator's game now. For a first-time borrower, the practical 2026 read is stark: either underwrite to Planet Fitness-style scale with capital to match, or pick a small-format boutique with a proven local niche, and treat anything in between as the category's documented kill zone.

Automotive: slow category, sharp exceptions

Automotive as a segment sits with QSR below half a percent growth (14), yet it contains some of the smartest deal structures in franchising. Christian Brothers Automotive discloses average store revenue around $2.7 to $2.8 million while holding the franchisee's Item 7 to $550,250 to $680,400, because the franchisor develops and owns the real estate and leases it back (35). The borrower gets a seven-figure business on a six-figure loan; the tradeoff is a profit-split model and less residual real estate upside. The loan record is spotless: zero charge-offs on 137 resolved cohort loans at a median $309,000 (2). The category's legacy repair flags are the other pole, with Meineke at 24.3 percent and AAMCO at 22.1 percent against an 11.7 percent franchised auto-repair average (2).

Take 5 Oil Change makes the format decision explicit: a conversion runs about $287,145 while a ground-up build runs to $2,053,642 (36). Same brand, same royalty, a sevenfold difference in borrowed money. Anyone comparing automotive franchises is really comparing entry formats. The mobile-tool franchises anchor the light end, with Matco at $104,374 to $376,241, a $10,000 fee and no royalty, though its 2025 churn, 281 closures against 211 openings, is the number to watch there (37).

Pet care: small loans against a $165 billion bowl

Pet spending reached $158 billion in 2025 and is projected at $165 billion for 2026, spread across roughly 95 million pet-owning households, with services the fastest-growing slice (38). Franchise entry runs $194,750 to $470,450 (21), squarely mid-ladder. The demand is demographic and sticky; the constraint, where veterinary care is involved, is clinical labor. Grooming, daycare, and boarding concepts carry the cleaner staffing profile and the smaller buildouts, which is what a lender wants to see against a discretionary-adjacent revenue line that has so far refused to act discretionary.

The caution list, handled like adults

Every ranking in this genre eventually publishes a worst-of table, and most of them do it badly: no cohort, no denominator, no acknowledgment that brands change hands. So here is ours, done properly. Every rate below is our own computation from the SBA loan-level file (2): fiscal 2010-2019 approvals, resolved loans only, minimum 40 resolved loans per brand, counts shown so you can check the math. The figures describe the loans in the file, under the flag they were originated under, and nothing more.

Brand	Charge-offs / resolved	Rate	Context
Experimac	43 / 65	66.2%	The defining attribution lesson: rebranded Experimax in 2019 after a trademark dispute and since largely wound down (40). Nobody can buy this loan history.
Window Genie	25 / 45	55.6%	Proof the booming home-services category still produces casualties at the thin-territory end.
Dental Fix Rx	23 / 44	52.3%	Mobile dental-equipment repair vans; a niche service line that never found its floor.
Burgerim	25 / 63	39.7%	The late-2010s collapse that helped trigger congressional scrutiny of SBA franchise lending (9).
Signarama	26 / 68	38.2%	Sign-shop retail squeezed by digital print.
The Grounds Guys	20 / 53	37.7%	Landscaping territories; same category as Window Genie, same lesson.
Orange Leaf Frozen Yogurt	20 / 54	37.0%	The froyo cycle in one line; Menchie's sits at 15.9 percent in the same file (2).
Tutor Doctor	21 / 58	36.2%	In-home tutoring, while the brick-and-mortar school brands top the honor roll.
Fantastic Sams	26 / 72	36.1%	Value haircare, against Great Clips at 1.8 percent. Category is not destiny.
Dickey's Barbecue Pit	51 / 155	32.9%	The one currently active risk story: reported state registration lapses in March 2026, a California consent order effective June 2026, and a $700,000 arbitration award over understated cost disclosures upheld in July 2026 (44)(45). An independent tracker on a narrower loan count lands near 29 percent (43).
iLoveKickboxing	27 / 94	28.7%	Boutique fitness at peak froth.
Which Wich	27 / 109	24.8%	The sandwich mid-market, squeezed between Subway's price and Jersey Mike's momentum.

Four names that dominate the older worst-of lists are missing from ours, and the reasons are the point. Planet Beach left only 17 resolved cohort loans, below our threshold, though trackers of the full 1991-to-present file put its longer history near 56 percent (39); its franchisor pivoted to an affiliated concept, HOTWORX, which now ranks second among the most-financed franchises of the past three fiscal years in the very same file, the strangest redemption arc in the dataset (2). Quiznos runs 19.5 percent on just 41 cohort loans against a 42.2 percent all-time figure (42), because the collapse from 4,700 units largely predates fiscal 2010. Beef 'O' Brady's posts 12.5 percent on 24 cohort loans against 54.9 percent all-time (39), two eras of one flag, with the current owner expanding and disclosing a $2.59 million new-unit run-rate (41). And Golf Etc left just 10 cohort loans behind a 54.3 percent legacy rate (39). Old lists memorialize old systems. The cohort tells you which risks are still breathing.

The pattern behind the table is older than any brand on it. The Wall Street Journal's 2014 pass over the same federal data found the ten worst chains responsible for $121 million in unpaid SBA-guaranteed loans, a fifth of all franchise write-offs (46). GAO found that roughly 28 percent of franchise 7(a) dollars from fiscal 2003 to 2012 ended in guaranty payments (8), and the Coleman Report was documenting a franchise default surge as far back as the 2008 crisis (47). High-default brands are not a 2026 anomaly. They are a permanent feature of a market where disclosure documents are written by sellers, which is precisely why the loan tape deserves a seat at the table.

One standing note: everything above is computed by us from a public federal dataset, and the underlying loan-level file is downloadable by anyone (2), as is our brand table at the end of this article. Brand operators who believe a figure misstates their current system are describing a real phenomenon, origination-era attribution, and the remedy is the one used throughout this piece: date-stamp the cohort, and read the current FDD alongside the old loans.

Financing the purchase in 2026

Assume the brand clears the Directory check. Here is the deal you are actually walking into this year.

Equity comes first. A startup or a complete change of ownership requires at least 10 percent of total project cost injected, and if part of that injection is a seller note, it counts only on full standby for the life of the loan (4)(11). On the median project midpoint of roughly $362,000 that is a $36,000 check before anything else moves (21); on a Culver's it is several hundred thousand. Pricing is the second reality: with prime at 6.75 percent, variable 7(a) loans are clearing at roughly 9.75 to 13.25 percent APR, and the average 7(a) loan sits near $448,000 (20). Real-estate-heavy formats, the drive-thrus and big-box gyms, routinely pair a 7(a) working-capital structure with a 504 debenture on the property, which is how a $3 million project gets financed without a $3 million single note.

October 1 raises the bar again. Under SOP 50 10 8.1 as issued in August, first-time acquisition borrowers face a reported 1.25x historical debt-service-coverage test, and deals at $3 million and above trigger a lender-ordered quality of earnings review (13). Translation for anyone buying an existing unit: the seller's tax returns, not the franchisor's Item 19, are the document that approves or kills your loan. Get three years of them before you get attached.

And the feasibility study question, which this site exists to answer honestly: no rule requires one. Nowhere in SOP 50 10 8 is there a sentence mandating a feasibility study for franchises, startups, or any asset class; the regulatory basis, 13 CFR 120.160(b), says only that SBA may require one (48). What actually happens is lender discretion doing its job. On projection-based credits, a first unit with no operating history, a ground-up build, a special-purpose property like a gym or drive-thru where the SOP separately mandates a specialized going-concern appraisal, lenders order an independent feasibility study because the credit memo needs a third party to test the revenue curve the borrower is betting on (4)(48). If your 2026 deal is a new unit of anything on this page, budget for one, choose the preparer before the lender chooses for you, and treat an independent feasibility study as the cheapest stress test your own money will ever get. Early-default rates on franchise loans have already ticked to roughly double their historical norm, and lenders have noticed (49).

The shape of the right answer

Strip away the brand names and 2026's best franchise is a shape: a low-buildout, demographically backed business in a category still adding units, carried by a brand whose seasoned charge-off record is clean and whose disclosed unit volumes cover 2026-priced debt with room to breathe. Home services, child care, and senior care fit the shape at small loan sizes. Culver's, the drive-thru beverage leaders, Planet Fitness at scale, and the smarter automotive structures fit it at large ones. The brands on the caution list, read with their dates on, mostly show what happens when the shape is violated: big obligations, thin volumes, and a disclosure document that promised otherwise.

The loan tape will keep score either way. It always has.

Frequently asked questions

What is the SBA loan default rate for franchises? By our calculation from the SBA loan-level file (March 31, 2026 release), franchise-flagged borrowers in the seasoned fiscal 2010-2019 approval cohort charged off 9.4 percent of resolved 7(a) loans versus 7.1 percent for non-franchised borrowers (2). An independent tracker on a later release lands at 10.2 versus 7.8 percent (1), and the FTC's narrower analysis found 3.9 versus 3.5 percent (7). The level depends on methodology; the franchise premium appears in every serious cut.

Which franchises have the highest SBA loan default rates? In our computation of the fiscal 2010-2019 cohort (minimum 40 resolved loans), the highest brand charge-off rates are Experimac at 66.2 percent, Window Genie at 55.6 percent, Dental Fix Rx at 52.3 percent, Burgerim at 39.7 percent, and Dickey's Barbecue Pit at 32.9 percent (2). Older all-time lists add Planet Beach, Beef 'O' Brady's, Golf Etc, and Quiznos at 42 to 56 percent (39)(42), but those rates largely describe pre-2010 loans under prior ownership, so always date-stamp the cohort before applying an old rate to a current brand.

Does the SBA publish default rates by franchise brand? No. The SBA publishes the raw loan-level 7(a) and 504 file with the franchise name on each record, updated quarterly, but no brand-level default table (2). Congressional proposals would require quarterly brand-level disclosure, but none has become law (9). Brand rates in circulation are computed from the public file by third parties.

Do I need a feasibility study to get an SBA loan for a franchise? No rule mandates one. The regulation says the SBA may require a feasibility study, and in practice lenders order one at their discretion on projection-based credits: new units, ground-up construction, and special-purpose properties (48)(4). If your deal has no operating history, expect the request.

What changed in SBA franchise lending for 2026? The Franchise Directory is back and mandatory (effective June 1, 2025), franchisors faced a June 30, 2026 certification deadline with removal for non-compliance, affiliation is now ownership-only, startups and full buyouts require 10 percent equity, the SBSS prescreen retired on March 1, 2026, and SOP 50 10 8.1 takes effect October 1, 2026 with a 1.25x acquisition coverage test and a $3 million quality-of-earnings trigger (4)(5)(6)(10)(12)(13).

How much money do I need to buy a franchise in 2026? Plan for at least 10 percent of total project cost in cash equity (4)(11). Median Item 7 investment bands run from $77,500-$152,100 in business services and roughly $118,000-$243,000 in senior care up to $358,500-$822,250 in food and beverage, with premium drive-thru and big-box concepts far above that (21).

August 30, 2026 by Michal Mohelsky, principal of MMCG Invest, LLC, a national SBA and USDA feasibility study consultancy

Request a Feasibility Study https://calendar.app.google/EJzWEz3GCqLY2jU86

Michal Mohelsky, J.D. | Principal | mmcginvest.com 

Contact: michal@mmcginvest.com

Phone: (628) 225-1125

Disclaimer: This report is provided for informational purposes only and does not constitute investment, legal, or tax advice. Data presented herein is derived from proprietary MMCG databases and third-party sources believed to be reliable; however, MMCG Invest makes no representation as to the accuracy or completeness of such information. Figures from third-party industry databases have been independently verified and, where appropriate, adjusted to reflect MMCG's proprietary analytical methodology. Statutory and regulatory references are provided for context and must be verified with counsel before reliance. Past performance is not indicative of future results.

Sources

(1) Franchise SBA Default Rates study, franchisefailurerates.com, FY2010-2019 approval cohort, loan statuses as of June 30, 2026.

(2) U.S. Small Business Administration, 7(a) and 504 FOIA loan-level dataset; all figures identified as our calculation are the author's computations from the March 31, 2026 release (resolved = paid in full + charged off; brand names normalized).

(3) International Franchise Association / FRANdata, franchise share of SBA lending (November 2025 release).

(4) U.S. Small Business Administration, SOP 50 10 8 (effective June 1, 2025); Information Notice 5000-868665.

(5) Taft Law, SBA Franchise Directory Reintroduced Effective June 1, 2025.

(6) NAGGL, Franchise Directory Registration Deadline Extension (to June 30, 2026).

(7) Federal Trade Commission, Franchise Issue Spotlight (SBA loan performance analysis).

(8) U.S. Government Accountability Office, GAO-13-759, Review of 7(a) Guaranteed Loans to Select Franchisees (2013).

(9) Restaurant Business, Senators push for more disclosure on SBA franchise loans.

(10) 13 CFR 121.301, size standards and affiliation principles for financial assistance programs, eCFR.

(11) Starfield & Smith, Best Practices: Franchise Lending under SOP 50 10 8.

(12) FRANdata, SBA Sunset of SBSS Scoring: What Franchisors Need to Know.

(13) U.S. Small Business Administration, Information Notice 5000-880695, Issuance of SOP 50 10 8.1 (effective October 1, 2026); acquisition provisions as reported from the notice.

(14) International Franchise Association, 2026 Franchising Economic Outlook (February 2026).

(15) Franchising.com, IFA Forecasts 12,000 New Franchised Units in 2026.

(16) FRANdata, 2025 Franchising Economic Outlook.

(17) Restaurant Business, The number of franchisee bankruptcies has soared this year (July 15, 2026).

(18) National Bureau of Economic Research, Working Paper 34033, Did California's Fast Food Minimum Wage Reduce Employment?

(19) Cato Institute, The Effects of California's $20 Fast-Food Minimum Wage on Prices (August 2026).

(20) Lumos Data, SBA 7(a) Loan Data and Program Performance: FY2026 Analysis.

(21) VetMyFranchise, Franchise Costs 2026: analysis of 2,185 FDD Item 7 filings.

(22) Culver's Franchise Disclosure Document (2026), Items 7, 19, 20; 2025 FDD for prior-year comparison. (state franchise registry)

(23) 7 Brew Franchise Disclosure Document (2025), Items 7 and 19. (state franchise registry)

(24) Scooter's Coffee Franchise Disclosure Document (2025), Items 7 and 19. (state franchise registry)

(25) QSR Magazine, Subway's U.S. Count Keeps Declining.

(26) Capital Digest, Subway Closes 729 More U.S. Stores in 2025.

(27) FRANdata, U.S. Franchising's Economic Outlook in 2026: Jobs, Output, and Growth.

(28) U.S. Census Bureau, American Community Survey 2024, median age of U.S. housing stock.

(29) Huntington Learning Center Franchise Disclosure Document (2025), Items 7 and 20.

(30) U.S. Census Bureau, National Population Projections.

(31) U.S. Bureau of Labor Statistics, Occupational Outlook Handbook: Home Health and Personal Care Aides.

(32) AARP, Home and Community Preferences Survey.

(33) Trade press reporting on Blink Fitness Chapter 11 (August 2024) and the Chapter 7 of the largest Orangetheory franchisee.

(34) Planet Fitness Franchise Disclosure Document (2026), Items 7 and 19.

(35) Christian Brothers Automotive Franchise Disclosure Document (2025), Items 7 and 19.

(36) Take 5 Oil Change Franchise Disclosure Document (2025), Item 7. (state franchise registry)

(37) Matco Tools Franchise Disclosure Document (2025/2026), Items 7, 19, 20. (state franchise registry)

(38) American Pet Products Association, 2026 State of the Industry Report.

(39) SBA Loan Data, Franchise SBA Loans and Default Rates by Brand, data as of June 30, 2026.

(40) Experimax, rebrand announcement: Experimac is becoming Experimax (2019).

(41) Beef 'O' Brady's Franchise Disclosure Document (2025), Item 19. (state franchise registry)

(42) Lopes Law LLC, Quiznos: SBA Loan Defaults at 42.2% Rate.

(43) Franchise Investor Data, Dickey's Barbecue Pit SBA loan performance (39 of 134 seasoned loans charged off).

(44) Restaurant Business, reporting on Dickey's state franchise registration lapses (March 25, 2026) and California DFPI consent order (effective June 1, 2026).

(45) Restaurant Business, federal court upholds $700,000 arbitration award against Dickey's over cost disclosures (July 17, 2026).

(46) Wall Street Journal analysis of SBA franchise loan default rates (September 2014), as summarized by Forward Franchising.

(47) CNN Money, SBA small business loan failure rate hits 12% (February 25, 2009, citing Coleman Report data).

(48) 13 CFR 120.160, Loan conditions (feasibility studies at SBA discretion).

(49) FRANdata, Creditworthiness as Strategy: The FUND Score Advantage (October 2025), early-default trend on franchise 7(a) loans.