The letter arrives with one sentence: we are unable to proceed with your application at this time, and we are not obliged to give reasons. Sometimes it arrives after two years of an existing relationship, with sixty days' notice. Founders read it as arbitrary. It is not. Banks make this decision through a documented process, and while the reasons vary by institution, the categories are predictable enough to design around.
The account is expected to receive money from strangers
The defining feature of a retail brokerage account is that money arrives from a large number of people the bank has never met, in small amounts, from many countries, by card and by transfer, and then leaves again. Every one of those inbound payments is a potential third-party deposit, a potential mule transaction and a potential fraud claim. The bank does not know the depositors, so it inherits your KYC quality. If your onboarding is weak, the bank is holding the exposure and it knows it.
This is why compliance teams ask for the AML programme, the monitoring rules and the vendor stack early. It is not paperwork theatre. They are deciding how much of your control environment they are being asked to trust.
Chargebacks and dispute exposure
Card-funded trading sits in a category where a customer who loses money has a strong emotional incentive to dispute the deposit, and dispute reason codes are broad enough that many of those claims get filed. Where a bank is also the acquirer, or is exposed through an acquiring relationship, it is carrying the contingent liability for refunds on a business that may not have the funds when the claims land. Elevated dispute ratios also trigger card scheme monitoring programmes, which bring their own costs and remediation obligations. Our articles on chargebacks and card approval rates cover that mechanism from the payments side.
Jurisdiction risk and correspondent pressure
Banks that move dollars, euros or sterling internationally do so through correspondent relationships, and those correspondents impose their own standards downstream. A bank that accumulates customers in high-risk categories and weakly rated jurisdictions risks questions from its correspondents, which is a far larger commercial problem than the fees from one brokerage account. That asymmetry, small revenue against large tail risk, is the whole of de-risking, and it is why entire customer categories get exited rather than individually assessed. The way FATF listings ripple through this chain is a good illustration.
Licence scope that does not match the client map
When a bank sees settlement data showing cardholders and account holders in countries the licensing entity cannot lawfully serve, the file changes character. It stops being a merchant with regulatory approval and becomes a merchant with a possible authorisation gap, which is a category most institutions will not hold. This is the reason that most often looks like something else when it arrives, and it is worth reading alongside which licences banks accept.
Ownership, substance and the address problem
A company whose registered office is a service address, whose directors are professional nominees, whose shareholders resolve into another holding company in a different opaque jurisdiction, and whose bank application is made in a country where nobody works, presents as a shell. Whatever the commercial reality, the file reads badly, and the compliance analyst writing it up has to justify the recommendation to a committee. Substance is not a marketing word here. It means staff, premises, tax registration and decisions actually taken in the country.
Sanctions exposure is the one item with no negotiation attached. If the client base, ownership or payment corridors touch sanctioned parties or territories, the application is over, and continuing to solicit those clients creates liability for the firm and its officers directly. Screening is a permanent operating control, not an onboarding step.
Small things that decide close calls
Several items rarely cause a decline on their own but push a marginal file over the edge. Marketing that promises outcomes, since a bank's reputational risk team reads your website. Bonus structures and aggressive incentives, which several regulators have restricted and which correlate with complaint volume. Affiliate traffic with no oversight, because you own the conduct of people you pay. Unclear separation between corporate money and client money. And an application that describes the business vaguely, since a compliance analyst who cannot write one clear paragraph about what you do will not recommend approval.
What actually changes the outcome
Fix the structural items first, because polish does not survive contact with a risk model. Get the licence and the client map into alignment and block the countries you cannot serve, at sign-up and again at deposit. Build the ownership chain to natural persons and be ready to evidence source of wealth. Put substance where you are applying. Bring documented expected volumes, corridors and dispute ratios rather than making the bank estimate them. Show the monitoring rules and the appointed compliance officer, and be honest about your dispute history, since a bank that discovers it later treats the omission as the problem.
Then accept that the stack will be plural. Most trading firms run a bank relationship for corporate and treasury, separate arrangements for client funds where segregation applies, and payment institutions for collection and payouts. That is normal architecture rather than a workaround, and it is described in the comparison of payout rails. SINGUARD builds the software these firms run on and holds no financial licence anywhere, so we watch this from the operations side rather than the banking side. The pattern is consistent: the firms that get banked are the ones whose story survives being written down in one paragraph by a stranger.
"Nobody at the bank hates your business. Somebody there has to sign a memo saying the account is inside appetite, and if your licence does not cover where your clients live, that memo cannot be written."
— Roman Onta, Executive Director, SINGUARD
Key Takeaways
- Banks inherit your KYC quality on every inbound retail deposit, so weak onboarding becomes their exposure.
- Dispute liability and card scheme monitoring make card-funded trading a costly category to hold.
- De-risking is an asymmetry: small fee income against correspondent banking risk, which is why whole categories get exited.
- Substance, a visible ownership chain and a licence matching the client map move more files than a polished application.
Frequently Asked Questions
Why would a bank close an account that has worked for two years?
Periodic review. Institutions re-rate customers on a cycle and after events such as a change in country risk ratings, a rise in disputes, a shift in payment corridors or new correspondent requirements. The account can move outside appetite without the firm doing anything differently.
Does a better licence guarantee a bank account?
No. The licence is one input. Ownership transparency, substance in the country of application, the payment mix and the coherence between licence and client base are weighed alongside it, and a weakness in any of them can still produce a decline.
Is using several payment institutions instead of a bank a red flag?
Not by itself. Most trading firms run a plural stack: a bank for corporate and treasury, regulated arrangements for client money where required, and payment institutions for collection and payouts. What matters is that each relationship is disclosed and used for what it is licensed to do.
About the Author
Roman Onta is an Executive Director at SINGUARD. He builds the Prop Firm CRM, the Broker CRM, Scalegram and CopySignals side by side with his brother Alex Onta, and he helped on the design of eTrader, the division Alex built and leads. His ground is worldwide payment processing, AML compliance and the corporate structures brokers are built on, work the two of them carry together, shaped by executive roles in the UAE and international corporates. He lives and works in Dubai for most of the year. Meet the executive duo leading Singuard's five divisions.