Most weeks I take intro calls that open with the same sentence: I want to launch my own FX business. Five minutes in, about half the people on those calls are describing a business they have never read a guide about. They have read plenty of guides. The guides were about something else.
Their clients are not traders. They are importers paying suppliers in Shenzhen, agencies collecting fees in three currencies, contractors waiting on a payout from Lagos. The person on the call is a consultant, an accountant, a company formation agent or a broker who has spent years earning those relationships, and every month the same question lands in their inbox: can you help us move this money. That is a payments business. Search the phrase they use, and the internet answers for a trading business instead.
Two businesses share one name
A retail FX or CFD brokerage sells access to price movement. It runs on a trading platform, a liquidity provider, a bridge, a CRM and a risk desk that decides which flow to internalise. Revenue comes from spread and commission on speculative volume, and the hard part is acquiring traders who churn. The costs are public and well documented: a lean offshore launch runs somewhere around $50,000 to $150,000 in year one, a CySEC-regulated build $350,000 to $800,000 before regulatory capital, and a fully licensed FCA operation past a million.
An FX payments business sells settlement. The client wants money to arrive, in the right currency, to a named beneficiary, with something they can attach to an invoice. It runs on accounts, corridors, conversion, onboarding checks on your own clients, a ledger that can answer who owns what at any moment, and a named person who owns compliance. Revenue comes from FX margin and transfer fees on commercial flows that repeat because the underlying trade repeats.
The second market is the larger one by a distance. Cross-border payment flows totalled $194.6 trillion in 2024 and are forecast to reach $320 trillion by 2032, with the B2B slice alone at $31.6 trillion and heading for $50 trillion, according to FXC Intelligence. None of that volume is speculative. It is companies paying other companies.
| Retail FX / CFD brokerage | FX payments business | |
| What the client wants | Exposure to price movement | Money to arrive |
| Core stack | Trading platform, liquidity, bridge, risk management | Accounts, corridors, conversion, ledger, compliance |
| Revenue | Spread and commission on trading volume | FX margin and fees on commercial flows |
| Client lifetime | Short, acquisition-driven | Long, tied to the client’s own business |
| First question from regulators | Who bears market risk | Whose money is it, and where is it |
How to tell which one you are being asked to build
The fastest diagnostic is the questions your clients already ask you.
If they name currencies and countries, they want payments. If they name leverage, spreads and platforms, they want trading. If they ask whether money can sit somewhere between the sale and the supplier payment, they want accounts. If they ask what happens when a payment is late and who they call, they want an operator, not a terminal.
On our calls the single most common request is a list of currencies and corridors. Nobody launching a CFD brokerage has ever asked me for that. Everyone with a book of commercial clients asks it in the first ten minutes, usually with two or three specific corridors already in mind, because a client has been complaining about one of them for months.
That request is also the point where the business case becomes concrete. A corridor that a broker on Currencycloud, Ebury or iBanFirst cannot serve today is a corridor their client is currently paying someone else to solve, badly.
What the payments version actually requires
Here is where most first attempts go wrong. The instinct is to solve each need with its own integration: one provider for accounts, another for the corridor that the first one does not cover, a third for cards later. The first integration is cheap and feels like progress. The second one is where the real cost appears, and it is rarely on the invoice.
Every provider you add brings its own onboarding flow, its own definition of a settled balance, its own reporting format and its own reconciliation. Add the second one and you have quietly started a reconciliation department. Your client sees one brand, yours, and expects one answer about where their money is. Behind that brand, three systems disagree with each other by design, and none of them is responsible for the total.
That is not a theoretical risk. Synapse, a middleware platform sitting between roughly a hundred fintechs and their sponsor banks, filed for bankruptcy in April 2024 after its ledgers stopped matching the banks’ records, and end users spent months locked out of their own balances while a trustee tried to work out who owned what. Seven sponsor banks took consent orders from US regulators between 2022 and 2025. The technology worked. The accounting between the parties did not, and no contract had made anyone responsible for the whole picture.
So the useful question to ask an infrastructure provider is not how good the API is. It is who reconciles, how often, and what happens when two systems disagree. One client record instead of three. One ledger that is the source of truth. Clearly defined compliance responsibilities across the products you offer. The ability to change a provider underneath without re-onboarding every client you have.
A word of warning on vocabulary, because the industry has muddied it. In 2026 orchestration usually refers to routing transactions between payment processors to lift authorisation rates. That is a real discipline and a different problem. What a business putting its own name on financial products needs is a layer above the providers that keeps identity, balances and accountability in one place.
The route that does not start with a licence
The good news is that the payments route has a first step that costs less than the trading route’s cheapest version.
Most partners we work with move through three stages. They start by introducing clients to a licensed operator and taking a share of the revenue those clients generate, which requires no build and no licence. When the flow is proven, they move their brand to the front: their name on the platform, their pricing, their client relationship, running on someone else’s permissions and infrastructure. Some eventually apply for their own licence, once volume makes the fixed cost of running one rational. That last step deserves a proper comparison of what an EMI licence and a banking licence each demand before anyone commits to it.
The honest limits are worth stating. Onboarding a white-label programme is a paid commitment, and providers ask for it because implementation consumes real work on their side. Compliance will decline some of your clients, and you need to know the criteria before you sell to them. Your provider’s risk appetite becomes yours the moment you put your brand in front of it. Moving between providers later costs money and client patience, which is exactly why the structure underneath matters more than the launch date.
This is the layer Framnex builds: branded banking, FX and payment products for businesses that already own the client relationships, with one contract, one ledger and clearly defined compliance responsibilities across the setup.
Five questions before you sign anything
Whoever you build with, take these into the conversation. Who holds the client record, you or the provider. Who reconciles balances, and how often. What happens to your clients when a corridor closes or a provider changes its risk appetite. Whose name appears on the statement your client receives. And what the second product costs you after the first one is live, in work as well as in fees.
The answers separate an infrastructure partner from a supplier. A supplier sells you a connection. A partner is still accountable when two ledgers disagree at month end, which, sooner or later, they will.
Alex Zhukov is a co-founder of Framnex, which builds financial infrastructure for businesses launching branded banking, FX and payment products.
