There is a particular conversation I have had too many times. A founder calls because their sponsor bank has just paused a feature, or asked them to hold more reserves, or in the worst cases, asked them to wind the program down. The founder is angry. They feel ambushed. They didn't see it coming.
When we walk back through how the relationship started, the same pattern shows up. The bank was picked because the BaaS platform recommended them. The interchange split looked competitive. The launch timeline looked fast. The founder never met the bank's BSA officer, never read the bank's exam report, never asked who the bank's other program partners were. They picked a bank the way you pick a vendor.
A sponsor bank is not a vendor. It is your operating license. The relationship will outlast most of your other vendor relationships, and when something goes wrong, the bank's worst day will become your worst day. The diligence on day one is what determines whether month eighteen is fine or feral.
This guide is the twelve-question checklist I use when I am sitting on the founder side of a sponsor-bank evaluation. None of the questions are clever. All of them are routinely skipped. Each one is included because I have seen what happens when the answer turns out, in hindsight, to be a problem.
TL;DR
- A sponsor bank is your operating license, not a vendor. Diligence accordingly.
- The regulator behind the bank matters as much as the bank itself.
- Ask for offboarding history. Ask for current program partners. Ask the BSA officer to take a meeting.
- Get reserves, holdbacks, and monthly minimums in writing with re-pricing triggers documented.
- Get the exit playbook in writing before you sign the entry document.
- Call at least one two-year program partner before signing. No exceptions.
Why this matters more than the product roadmap
When founders compare sponsor banks, they tend to spend most of their time on UI, API quality, and dashboard features. Those things matter operationally, but they are not what determines whether the relationship survives. The relationship survives or breaks on three things: regulatory posture, governance discipline, and economic alignment. None of those are visible in a product demo.
A sponsor with a beautiful dashboard and a fragile BSA program will offboard you in year two. A sponsor with an ugly dashboard and a stable supervisory record will be there at year five. The dashboard is the second-order question. The first-order question is whether the bank's own regulator is comfortable with the program.
The shape of the conversation
When you walk into a sponsor-bank diligence conversation, the bank's BD team will try to keep the discussion on product features, customer success stories, and roadmap. That is appropriate from their side and unhelpful from yours. Your job is to redirect to the questions that matter for the life of the relationship.
A useful framing: you are doing diligence on the bank in the same way the bank is doing diligence on you. They will ask about your beneficial ownership, your funding, your KYC procedures, your projected volumes, your prior compliance history. You should ask about their regulator, their exam history, their program-management staffing, their offboarding pattern, and their reference customers. The conversation should feel symmetric. If it does not, that is information.
The questions, with notes
Regulatory posture (1–4)
The first four questions surface the regulatory environment the bank is actually operating in. A bank's primary federal regulator (question 1) is the single most consequential fact about it from a fintech-program perspective. The OCC, the FDIC, and the Federal Reserve have meaningfully different cultures, different exam tempos, and different recent histories with fintech programs. Knowing which one your bank reports to tells you a lot about the environment your program will live in.
Offboarding history (question 2) is the most diagnostic question I know. A bank that has offboarded one or two programs in the last 24 months and can describe each thoughtfully is a bank that has matured its third-party risk function. A bank that has offboarded none is either inexperienced or being selective with what it shares. A bank that has offboarded ten is one whose own risk posture is unstable.
Durbin status (question 3) is the largest single economic variable. If your sponsor is exempt today and on track to cross the threshold in 18 months, you should be modeling the post-graduation economics, not the today economics, because that's the world you will be operating in when your contract auto-renews.
MRAs and MRBs (question 4) are confidential, but the meta-conversation is not. A sponsor whose governance team will not engage on this dimension is one whose own posture is brittle. You don't need the specific findings. You need to know whether the bank thinks about its own regulator on a daily basis.
Operational depth (5–6)
The BSA officer and the third-party risk officer (question 5) are the two people whose decisions will define your program's life. Meeting them before you sign is the cheapest insurance you can buy. If the bank tells you those leaders are too busy to meet a prospective program partner, that is the level of attention you will receive after signing.
Program-launch timeline (question 6) is a maturity test. A bank that has done this twenty times will give you a specific timeline with named bottlenecks. A bank that gives you a vague answer is one whose process is still being invented around you. Both are workable; only one is honestly priced.
Economics (7–10)
Interchange splits (question 7), FBO ledger reconciliation (question 8), exit terms (question 9), and the reserves/holdbacks/minimums triad (question 10) are the operational economics of the relationship. Every one of them should be in writing before you sign. Every one of them should have a defined re-pricing mechanism with notice periods and dispute procedures. A bank that refuses to put any of these in writing is reserving the right to change them unilaterally, and they will.
I want to underline question 9 specifically. Model the exit before you model the launch. Every sponsor relationship ends. You may end it (you outgrow them, you graduate to direct issuance, you change strategy). They may end it (regulatory pressure, risk-appetite change, internal restructure). The cleanest exit is the one that was documented at signing. The most expensive exit is the one negotiated under duress.
The reality test (11–12)
The governance cadence question (11) tells you how the bank will treat you when something goes wrong. The reference-customer question (12) tells you what it has actually been like to be a partner for the long haul.
I want to be specific about question 12. Ask for a two-year program partner, not a six-month one. Six months in, every relationship looks fine. Two years in, every relationship has been tested at least once. The story of that test is the story you are buying into. If the bank cannot or will not make the introduction, that is the most informative answer in the entire diligence process.
How to use this checklist
Walk into your sponsor-bank diligence with the twelve questions printed on a single sheet of paper. Use them to score every candidate. Lock the scoring before any sales call. The sales process will try to reframe what matters; the checklist prevents that.
By the end of the diligence process you should have, in writing:
- A documented regulatory posture (regulator, exam tempo, current open findings to the extent shareable)
- A documented set of operational contacts (BSA officer, TPR officer, relationship manager) with whom you have spoken
- A documented interchange split, reserve methodology, holdback policy, and monthly-minimums structure
- A documented exit playbook including BIN portability
- Two reference conversations, including at least one with a two-year partner
If any of those is missing, you do not yet have enough information to sign. The cost of running this diligence properly is two to four weeks. The cost of running it badly is two to four years of an operating relationship you cannot escape gracefully.
Closing thought
A sponsor bank is not a feature comparison. It is a long-term institutional partnership in a regulated industry, with consequences that compound on a multi-year horizon. The most expensive mistake in fintech is treating it like anything less.
If you want to think about this in the broader frame of how Chisel sees fintech infrastructure, the Software as a Relationship pillar lays out why we believe direct, named, contractual relationships are the durable asset class in this industry. The sponsor-bank relationship is the foundational instance of that idea. Choose it like everything depends on it. Operationally, everything does.
— Darin Petty