# Who really controls the customer relationship
Open the Cash App on your phone and you are looking at a product built by Block, Inc. But the money you deposit is not actually held by Block. It sits at Sutton Bank or Cash App Bank (formerly Wells Fargo predecessors, now Block's own industrial bank charter), it moves through payment rails licensed by Visa and the ACH network (Automated Clearing House, the US batch payment system), and much of the account infrastructure behind it may be run on technology from a core banking vendor like Galileo (also owned by Block) or Marqeta.
None of that plumbing is visible to you. You think you bank with Cash App. That perception is the whole ballgame.
Let's mapmapUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.Voir la définition complète → the same $1,000 deposit through three different arrangements.
The Tier-1 bank (e.g., JPMorgan Chase, Bank of America). It owns the charter (the legal license to hold deposits and lend), the balance sheet, the brand, the app, and the branch network. It controls the entire stack, and it captures the entire relationship: your paycheck, your mortgage, your customer service calls. Full vertical integration.
The core banking vendor (e.g., Galileo, Marqeta, Fiserv). These companies do not talk to you at all. They sell infrastructure, APIs (application programming interfaces, the technical connectors that let software systems talk to each other), and ledger systems to *other* companies who build consumer products on top. Galileo's customer is Cash App, Chime, or Robinhood, not the depositor.
The neobank/fintech app (e.g., Cash App, Chime, Varo). It owns the brand, the app, the user experience, and the customer data. It does not own the charter. It rents balance sheet capacity from a partner bank under a "banking-as-a-service" (BaaS) arrangement, a model where a licensed bank provides the regulatory chassis while a fintech provides the front end.
Same depositor, three completely different power positions.
Here is the counterintuitive part: the entity with the banking license (the Tier-1 bank, or the small partner bank behind a neobank) often has *less* leverage than the entity that owns the app.
Why? Because depositors are loyal to the interface, not the plumbing. Most Chime customers could not name Chime's partner banks (The Bancorp Bank and Stride Bank). If Chime switched sponsor banks tomorrow, almost no customer would notice or leave. But if Chime's app had a bad review score or a clunky sign-up flow, growth would stall immediately.
This is a classic pattern across tech-enabled industries: the layer closest to the customer captures the brand equitybrand equityThe commercial value your brand adds beyond functional product attributes: the price premium, preference and loyalty it generates.Voir la définition complète →, the pricing power, and the data, even when it does not own the regulated asset underneath. Compare it to how Uber does not own cars, and Airbnb does not own real estate.
Sponsor banks like The Bancorp Bank or Cross River Bank are profitable, but they are largely invisible and substitutable. Fintechs have, in several documented cases, switched sponsor banks with minimal customer disruption. That substitutability is precisely what keeps the balance of power tilted toward distribution.
Power is not just commercial, it is also regulatory, and here the small partner banks matter enormously even if customers never see them.
In the US, the actual deposit-taking entity is supervised by the FDIC (Federal Deposit Insurance Corporation) or the OCC (Office of the Comptroller of the Currency), and it bears the compliance burden for BSA/AML (Bank Secrecy Act / Anti-Money Laundering rules). When something goes wrong, regulators go after the chartered bank, not the app.
This happened in 2024 when the OCC and FDIC took enforcement actions against several BaaS sponsor banks for weak oversight of fintech partners, and again when the collapse of banking-as-a-service middleware provider Synapse in 2024 froze deposits for end users of multiple fintech apps, exposing how fragile the "invisible" layer can be when things break. Customers of fintechs like Yotta discovered, painfully, that "your money is FDIC-insured" claims depend entirely on which entity actually holds the ledger of record, and reconciliation between a fintech's records and a bank's records is not automatic.
For a plain-language regulatory explainer, the Consumer Financial Protection Bureau's guide to bank partnerships is a useful primer on how these arrangements are supervised.
In Europe, the equivalent structure exists under the EMI (Electronic Money Institution) license regime, regulated nationally (e.g., by the FCA in the UK or BaFin in Germany) but passportable across the EU under PSD2 (the second Payment Services Directive). Revolut, for example, operated for years as an EMI before obtaining a full banking license, meaning its early "deposits" were technically e-money, safeguarded but not deposit-insured the same way as bank deposits.
This is where the module's core question, how value is distributed across the chain, becomes concrete.
Roughly, in a typical BaaS-powered neobank stack (figures are industry estimates, not disclosed exact splits):
The uncomfortable truth for infrastructure players: Galileo and Marqeta are essential, but they are also replaceable and price-competed. Chime has historically used Galileo; if Galileo's pricing became uncompetitive, Chime has both the scale and incentive to build in-house or switch, which is exactly what larger fintechs increasingly do once they hit sufficient volume, sometimes called "graduating" off vendor infrastructure.
Meanwhile, a Tier-1 bank like JPMorgan sits outside this whole dynamic. It does not need a BaaS vendor or a sponsor relationship because it owns the full stack. Its power comes from scale, trust built over a century, and direct regulatory relationships, but it is also the slowest mover and the most expensive to build new digital products on.
Vérification des acquis
1. In the Cash App example, why does the average user believe they 'bank with Cash App' even though a chartered bank holds the actual deposits?
2. What is the key structural difference between a Tier-1 bank like JPMorgan Chase and a neobank like Cash App?
3. Why does a core banking vendor like Galileo occupy a fundamentally different position from both the Tier-1 bank and the neobank?
4. Select ALL correct answers describing what a 'banking-as-a-service' (BaaS) arrangement provides to each party.
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers that explain why 'the same $1,000 deposit' can represent three completely different power positions depending on the player.
Sélectionnez toutes les réponses correctes.
Three things follow from this structure, and they define competitive strategy in the sector.
1. Fintechs race to own the interface, not the balance sheet. This is why Cash App, Chime, and Revolut invest heavily in app design, rewards, and brand marketing rather than in obtaining banking licenses early. The license is a cost center you outsource until scale justifies internalizing it.
2. Core banking vendors compete on reliability and speed to market, not brand. Galileo, Marqeta, Unit, and Treasury Prime are essentially B2B (business-to-business) infrastructure companies fighting for developer mindshare and enterprise contracts, invisible to the public but critical to the ecosystem's plumbing.
3. Tier-1 banks defend by acquiring distribution or building their own digital brands. JPMorgan's Chase UK digital bank and Goldman Sachs's now-wound-down Marcus consumer effort were both attempts to compete on the interface layer using an incumbent's balance sheet advantage. Goldman's retreat from Marcus in 2023-2024 is itself a lesson: owning the charter does not guarantee you can win the interface game against nimbler product teams.
🎬 [VIDEO: "How Banking-as-a-Service Actually Works" - youtube.com/@a16z - a clear breakdown of the BaaS stack and why sponsor banks, middleware, and fintech apps split the value chain the way they do]