Executive Summary: Franchise lending looks like small business lending with a recognizable logo attached, and that resemblance is what gets underwriters into trouble. You are advancing capital to a franchisee entity, but the repayment probability on that file is partly determined by a second party you are not underwriting: the franchisor, whose litigation posture, termination practices, and transfer restrictions sit upstream of your collateral. Court records are the cheapest way to see both entities at once, provided you know which names to search and you are honest about which jurisdictions your data actually covers.
Why does franchise lending have a two-entity problem that generic underwriting misses?
Standard small business underwriting assumes the borrower controls the business. In franchising, that assumption is only partly true. The franchisee owns the entity, the equipment, the lease, and the receivables. The franchisor owns the brand, the operating system, the supply terms, the territory grant, and in most agreements the right to approve or block a transfer of the business. Underwrite only the franchisee and you have priced half the risk.
What does the franchisee actually own?
A franchisee owns an operating entity with a wasting license attached. The franchise agreement has a term, renewal conditions, and default triggers that the franchisor defines and enforces. If that license terminates, the franchisee still holds a lease on a build-out configured for one brand, equipment specified by that brand, and a customer base that walked in because of a sign the franchisee no longer has the right to display. The same equipment package is worth materially different amounts depending on whether the franchise relationship survives, which is why the credit question is not only "can this operator service the debt," it is "will this operator still be permitted to run this business for the life of the note."
Why does the franchisor's health show up in your loss rate?
Franchisors fail, and when they do, the franchisees they supplied do not fail politely one at a time. The TGI Fridays Chapter 11 filing is a useful public example: the chain filed in November 2024 after roughly 100 closures in the preceding year, and by April 2025 it was down to about 85 US restaurants from 233 at the end of 2023.[1][2] Its UK master franchisee filed for administration in the same period. A franchisee operating a profitable unit inside a collapsing system inherits supply interruption, marketing fund evaporation, and a resale market that prices the unit as if the brand were already gone.
The 2024 and 2025 wave of restaurant filings was broad rather than idiosyncratic, covering Red Lobster, BurgerFi, Buca di Beppo, World of Beer, and others, and legal analysts attributed the surge to cost structure and lease overhang rather than one-off management failure.[3] If your portfolio has brand concentration you did not measure, that concentration is a correlated exposure hiding inside a set of files that each looked independent at origination.
What do the 2024 through 2026 franchise performance numbers actually say?
Franchising as a sector is growing, and franchise credit performance is deteriorating at the same time. Both statements are true, and holding them together is the correct underwriting posture.
The International Franchise Association projects franchise establishments rising from 832,521 to roughly 845,000 units in 2026, a 1.5% increase, with output above $920 billion and employment near 8.9 million.[4] Unit growth is not the same as unit health. Sector expansion driven by new franchise sales tells you the franchisor's development pipeline is working, not that the operators inside it are servicing their debt.
What changed in SBA franchise credit performance?
FRANdata reported that early loan defaults, meaning businesses failing within roughly the first 18 months, surged by 213% over an 18-month window, with the early default rate reaching 1.4% against a historical range closer to 0.6% to 0.8%.[5] Broader SBA default rates climbed to 3.7% in 2024, the highest reading since 2012.[6] Early defaults are the most diagnostic category for franchise lenders because a business that fails inside 18 months rarely failed because of a macro shock. It failed because the unit economics were never there, or because the relationship with the franchisor broke.
What does brand-level failure do to an otherwise healthy franchisee?
It compresses the exit. A franchisee under stress normally has three ways out: refinance, sell the unit to another operator, or negotiate with the franchisor. Brand distress closes two of those doors at once. Buyers disappear, and the franchisor loses both the incentive and the capacity to work out a soft landing. What you observe is a borrower who was current for six quarters and then stopped, with no explanation anywhere in the borrower's own entity.
Why is franchisor-franchisee litigation a system-health signal rather than only a borrower signal?
Most underwriters treat a lawsuit involving the borrower as a borrower-level negative. In franchising, the more informative reading is often directional: who is suing whom, and how often.
What does a franchisor suing its own franchisees tell you?
A franchisor that sues a single franchisee for unpaid royalties is enforcing a contract. A franchisor that has sued twenty of its own franchisees in three years is running a system where litigation has become an operating tool, and that pattern usually reflects one of two conditions: the unit economics are failing broadly enough that royalty non-payment is systemic, or the franchisor is monetizing terminations and resales rather than unit performance. Either condition raises the probability that your specific borrower will find themselves on the receiving end of a termination notice regardless of how well they run their store.
Franchise disclosure rules partially surface this. Franchisors must disclose material civil actions, franchise relationship actions, prior actions within a ten-year lookback, and government injunctions and orders in Item 3 of the Franchise Disclosure Document, including actions the franchisor brought against franchisees where the volume crosses the disclosure threshold.[7] That disclosure is genuinely useful.
Where does FDD Item 3 fall short for a lender?
Item 3 is a point-in-time document prepared by the counterparty whose behavior you are trying to assess, refreshed annually, and framed around what the franchisor is required to disclose rather than what you need to know. It will not show you the case filed last month. It will not show you the franchisee-side pattern in your specific market. It says nothing about your borrower's own litigation, judgments, or judgment creditors.
Regulators have been circling the gap between what franchisors disclose and how franchisors behave. In July 2024, the FTC issued a policy statement and staff guidance stating that contract terms restricting franchisees from communicating with government agencies are unfair and unenforceable, and that franchisors may not impose or collect fees that were never disclosed.[8][9] The guidance followed more than 2,000 public comments the FTC gathered on franchisor business practices. A disclosure regime that needs that kind of corrective guidance is not a regime you should treat as a complete risk picture.
Which franchise dispute types should actually change your credit decision?
Not every franchise lawsuit is a credit event. These categories are the ones that move the decision, because each one attacks either the borrower's right to operate or your position in the collateral.
Which dispute categories carry credit weight?
• Termination and cure disputes. The franchisor moves to terminate, the franchisee argues the default was cured or never occurred. Franchisee counsel routinely litigates terminations that followed a timely cure, seeking reinstatement, damages, or both.[10] A pending termination action against your borrower is a direct threat to the going-concern assumption in your credit memo.
• Encroachment and territory disputes. Where a franchisee holds exclusive territory rights, a franchisor opening a company-owned or franchised outlet inside that territory, or granting an overlapping grant to another franchisee, is a live claim.[11] Encroachment shows up in your file as revenue decline that the borrower cannot explain in operational terms.
• Post-termination damages exposure. Recent decisions have been read as favorable to franchisors on lost future royalties following a total breach, which means a terminated franchisee can carry a damages claim well beyond the unpaid balance at termination.[12] That claim competes with you.
• Fee and chargeback disputes. Technology fees, marketing fund assessments, and required remodels that the franchisee disputes are small individually and corrosive in aggregate, and they are precisely the category the FTC flagged in 2024.[9]
• Judgment and lien activity against the franchisee entity or its owners. This is the ordinary credit signal, and it is the one that most often exists in court records while being absent from every document in the application package.
How much should a court check cost before it is worth running on every file?
That question, not data quality, is what determines whether court records enter an underwriting flow at all. Gate Rock Capital put the threshold in explicit terms: "where I would pay $4 a pull is when you have the state index on court search." At that price a check is a research tool used on exceptions. At parity with an SOS lookup it becomes a standard field on every application, which is a different product decision entirely.
How do transfer-of-ownership restrictions affect your collateral position?
This is the part of franchise lending that most often surprises lenders who came from independent small business credit. Your borrower cannot freely sell the business, and in many agreements neither can you, in practice, if you take it back.
What does the franchise agreement actually restrict?
Franchise agreements typically require franchisor consent to any transfer of the franchise, grant the franchisor a right of first refusal, and impose approval standards on the buyer. That means the realizable value of your collateral in a default depends on a third party's consent decision. If the franchisor declines to approve a transferee, your enterprise-value recovery collapses toward liquidation value on used equipment and an assignable lease.
What did the SBA franchisor certification change in 2025?
The SBA sunset its Franchise Directory in 2023 under SOP 50 10 7, shifting the burden onto lenders to determine that the applicant met loan program requirements.[13] It reversed course effective June 1, 2025 under SOP 50 10 8, reinstating the Directory with a new mechanism: instead of executing Form 2462, franchisors execute a Franchisor Certification affirming compliance with the eligibility conditions for listing, with a July 31, 2025 deadline for previously listed brands.[14][15] By signing, the franchisor agrees not to enforce contract provisions inconsistent with the certification, which includes provisions bearing on the lender's ability to secure the loan with collateral.[16] Lenders now carry a verification checklist obligation confirming the brand is listed.[17]
The certification improves your position on paper. It does not tell you whether this particular franchisor litigates against franchisees, blocks transfers, or is itself financially distressed. Directory listing is an eligibility gate, not a credit opinion.[18]
Why does the franchisee's legal name almost never match the name on the application?
Here is the operational problem that defeats most franchise court searches before they start. The application says "Jersey Mike's Subs, Northport." The trade name on the storefront says the same thing. The entity that signed your note is something like "NPT Sandwich Ventures LLC." The judgment against that entity, if one exists, is indexed under the legal name, not the trade name, and searching the brand name in a court index returns the franchisor's litigation, other franchisees' litigation, and trademark matters, none of which are your borrower.
How do you resolve the legal entity before you search?
Secretary of State data is the resolution layer, and it has to run first. A live SOS pull gives you the registered legal name, entity status, formation date, registered agent, and in most states the officer and member records. That officer data is the second half of the search set, because franchise financing is very often personally guaranteed, and the guarantor's individual litigation history is frequently more informative than the entity's. A two-year-old LLC formed to hold one franchise unit has no litigation history by construction. Its owner may have plenty.
The sequence that works is mechanical:
• Resolve the legal entity from the trade name. Pull SOS by business name and confirm formation date, status, and address against the application.
• Extract the officer and member names. These become individual search terms for the court pass and for guarantor review.
• Check for related entities under the same officers. Multi-unit franchisees commonly hold each location in a separate LLC, and the litigation may sit in a sibling entity rather than the borrower.
• Run court records against the resolved legal name. Not the trade name, not the brand.
• Run court records against each guarantor individually. Judgment history follows the person across entities.
• Separately assess the franchisor's litigation posture. This is a different search with a different purpose, and its output informs brand-level concentration limits rather than the single file.
Skipping the resolution step produces a clean court report that is clean because it searched the wrong string. That is worse than no report, because it creates documented false comfort in the credit file.
Which names belong in the search, and which do not?
Search the franchisee legal entity, every guarantor by individual name, and any sibling entities sharing officers. Do not treat brand-name hits as borrower hits. A search for the brand in a busy jurisdiction will return dozens of unrelated matters, and an analyst under time pressure will either dismiss the whole report or escalate a case that has nothing to do with your applicant.
"If courts are cheap enough, then it's worth it to run on every application automatically." That is Yehudah Aron at Cucumber Capital, and it captures the real constraint. The blocker on court checks in franchise files is almost never whether the data helps. It is whether the check is cheap enough and fast enough to run on files that have not yet been approved.
What are the practical options for running the court search, and where does each one break?
Before naming any single tool, the honest framing is that franchise lenders already have working options and most of them are adequate for research and inadequate for automation.
Manual courthouse and clerk portal search is the baseline and is genuinely free or near-free. New York's Unified Court System offers NYSCEF case search for e-filed civil matters plus the WebCivil portals, searchable by index number, party name, or attorney.[19] Miami-Dade's Clerk provides free online case search across civil, family, and probate records, with modest per-name fees for advanced searching.[20] If you underwrite ten files a week, portal search works. Unicourt and LexisNexis offer aggregated multi-jurisdiction reach that no single-source option matches, at a price point suited to research and litigation support. PACER covers federal matters, which matters for bankruptcy but not for the state-court judgments that dominate small-business credit. CSC and Wolters Kluwer serve the corporate-services side, particularly where UCC and registered-agent work already sits with them.
The gap those options share is not coverage. It is integration friction. None of them drops cleanly into a decisioning workflow where a court check has to complete inside the same automated pass as the SOS pull, without a human opening a browser tab. Max Weisz, running an MCA shop at roughly 500 files a day, described the current state plainly: "we run New York court separately and then we run Unicourt, and then we run UCC searches." That is three tools and a person stitching them together on every file.
What does Cobalt's Court Case API actually cover, and what does it not?
State the limit first, because it determines whether this is useful to you at all. Cobalt's court records coverage is New York State and Miami-Dade County, Florida. That is the entire footprint. It is not nationwide, not federal, and not a PACER wrapper. If your franchise book is concentrated in Texas or the Midwest, this product does not solve your problem, and you should keep using the aggregators.
The coverage is demand-driven rather than aspirational. Roughly 80% of Cobalt's funder customers file judgments in exactly those two venues: New York because it is the center of alternative lending, and Miami-Dade because a large share of funders relocated to South Florida during and after 2020. Broader jurisdictions are on the roadmap, including a planned human-assisted queue for unsupported venues, which would run slower, around an hour, and is not yet shipped.
Prospects say the quiet part out loud, and the objections are fair. Lara Hodgson at RoxWrite put it directly: "Most of our clients are not in New York." Cameron Kelliher at Elementix reached the same conclusion and acted on it: "We'll stay away from the court stuff then. I wish the court stuff was rounded." Both responses are correct for their books. A franchise lender whose applicant base sits outside these two venues should not buy a court layer that cannot see its files, and no amount of API quality changes that arithmetic.
The endpoint is asynchronous. You post a lookup with a callback URL and receive results when the live court pull completes, typically in 30 to 120 seconds. There is no synchronous mode and there is no monitoring or alerting product.
curl -G "https://apigateway.cobaltintelligence.com/courtCases" \
-H "x-api-key: $COBALT_API_KEY" \
--data-urlencode "businessName=NPT Sandwich Ventures LLC" \
--data-urlencode "jurisdiction=newYork" \
--data-urlencode "callbackUrl=https://underwriting.example.com/hooks/court"Jurisdiction accepts `newYork`, `miamiDade`, and the test values `testNewYork` and `testMiamiDade`, which return sample payloads without consuming credits. Results include judgment details, case number, case type and division, filing dates, parties, and amounts where the record carries them; not every court record includes an amount. Pricing is one credit per lookup, the same unit cost as an SOS pull, which is what makes the run-on-every-file pattern arithmetically possible rather than a budget conversation.
Data is pulled live from the court site at request time rather than served from a cached index, which is the reason for the asynchronous design and also the reason a filing from last week is visible.
Running franchise files in New York or Miami-Dade and want to see what a live court pull returns before committing anything? Book a walkthrough at cobaltintelligence.com/lp/demo/ and bring three real applicant names.
How should a franchise file be sequenced end to end?
Court records are one layer, not the decision. The order matters because each layer produces the input for the next.
What is the correct order of checks?
• Start with SOS. Resolve the trade name to a legal entity, confirm active status, capture officers and members. All 50 states, live pull. This is the only step that makes the rest of the chain accurate.
• Run UCC next. Existing blanket liens tell you whether you are in first position before you spend money on anything else. Cobalt returns UCC data in the same call as SOS across roughly 10 to 11 states, billed as a separate credit.
• Run court records on the resolved entity and each guarantor. New York and Miami-Dade only, one credit per lookup, asynchronous with a callback.
• Verify TIN and EIN. All states, and it closes the identity loop between the entity you searched and the entity you are funding.
• Screen OFAC. Global coverage, and it is a compliance requirement independent of credit quality.
• Assess the franchisor separately. Pull the current FDD, read Item 3, check whether the brand is on the reinstated SBA Franchise Directory, and search the franchisor's own litigation in whatever venue you can reach.
Why is New York the strongest case for a single integration?
New York deserves a specific note: it is the only state where Cobalt's SOS, UCC, contractor licensing, and court records coverage all overlap. For a New York franchise file, that means one integration produces registration status, lien position, license standing where applicable, and judgment history. For a file in most other states, it produces the first and the last of those and you will need something else for the rest.
Position this correctly inside your stack. Cobalt is a data source, not a decisioning engine. It returns what the court and the state actually hold, at the moment you ask, in a format your underwriting system can consume. What that data means for a specific franchise file is a credit judgment, and it belongs to your team. For related reading on structuring that judgment, see our guides on building litigation risk scoring with court data, court records for SBA lending compliance, and detecting repeat defendants in lending applications.
Ready to add a court layer to your franchise underwriting in New York or Miami-Dade? Start at https://cobaltintelligence.com/lp/demo/.
.jpg){kind=small}
.png)