Choosing a licensing regime: a comparison framework
Start with your markets and your payment rails. The fee table is the least decision-relevant information available.
Most licensing comparisons start with a table of state fees and minimum capital. That table is the least decision-relevant information available, because those are the two numbers that vary least in their impact on whether the business works.
The wrong question
“Which licence is cheapest and fastest” produces a predictable outcome: an inexpensive authorisation that your payment providers will not accept, in a jurisdiction your target market does not recognise, obtained quickly and useless slowly.
The right question is narrower and harder. Given the markets we intend to serve and the payment rails we need, which authorisations are acceptable to every party whose acceptance we require — and of those, which can we actually maintain?
Note the two halves. Acceptability is about the outside world. Maintainability is about you: whether you can staff the key function holders, meet the reporting cycle and fund the capital requirement on an ongoing rather than a one-off basis.
The order the decisions actually come in
- Markets. Where will your customers be, in eighteen months rather than at launch? This is the only input that cannot be changed cheaply later.
- Activity classification. What exactly are you doing, in the regulator's vocabulary? A single product often spans several defined services, and the classification determines everything downstream.
- Payment rails. Which banks, acquirers and PSPs do you need, and which authorisations do they accept? Ask them before you apply, not after. This step is skipped in almost every project that goes wrong.
- Candidate regimes. Only now does the shortlist appear — and it is usually short.
- Maintainability. For each candidate: staffing, substance, reporting burden, ongoing capital.
- Cost and time. Last, because by this point there are rarely more than two viable options.
A comparison framework
Score each candidate regime against these dimensions rather than against a fee table. The weightings differ by business; the dimensions do not.
| Dimension | What to establish | Why it decides |
|---|---|---|
| Market coverage | Which target markets the licence lawfully covers, and which require separate local authorisation | A licence that does not cover your main market is a cost, not an asset |
| Counterparty acceptance | Whether your shortlisted banks, acquirers and software suppliers accept it | The most common reason a cheap licence turns out to be worthless |
| Substance required | Directors, key function holders, office, local residency, and when each is required | Determines your real running cost and whether you can staff it at all |
| Supervisory intensity | Reporting frequency, audit cycle, inspection practice, appetite for enforcement | Decides how much management attention the licence consumes every quarter |
| Ongoing capital | Not the entry threshold — the formula that applies once operating | Own funds tied to overheads or outstanding balances can exceed the entry figure quickly |
| Change flexibility | How new services, new shareholders or a new director are approved | A regime that makes change slow constrains the business, not just the paperwork |
| Exit and transferability | Whether the licence survives a change of control, and how long approval takes | Relevant the day someone wants to buy you, which is the day it is too late to check |
Costs nobody puts in the table
- Key function holders. A compliance officer and an MLRO who are genuinely available and genuinely independent are a recurring salary cost, not a line item at application. Several regimes now prohibit combining these roles with operational management.
- Substance. Registered office, local director, in some regimes a physical office on a phased timetable. The Curaçao regime, for instance, requires a resident managing director from the day the licence is granted and phases in further key persons and premises over subsequent years.
- Audit and certification. Financial audit, compliance audit, and in gaming an independent systems review against the live environment.
- Reporting. Someone has to produce the returns. In a small firm that someone is usually a founder, and the cost is measured in attention rather than in fees.
- The second application. If the first regime turns out to be unacceptable to your payment partners, you pay for two. This is common enough to be worth pricing as a risk.
A worked example
A crypto exchange intending to serve retail customers across the EU, with card acquiring for fiat on-ramp.
- Markets: EU retail, so a passportable EU authorisation rather than a third-country regime.
- Classification: exchange of crypto-assets for funds and execution of orders — and, if an order book is operated, a trading platform, which changes the capital class and adds market abuse obligations.
- Rails: card acquiring for crypto purchase is the binding constraint. Acquirers have narrow appetite, and their requirements should be established before the member state is chosen.
- Candidates: CASP authorisation in a member state whose supervisor has processed comparable applications and whose banking environment serves authorised CASPs.
- Maintainability: own funds at the higher of the class minimum or a quarter of fixed overheads — which for a firm with real staff costs will exceed the headline figure — plus effective management in the member state and an EU-resident director.
- Only then does cost and timing enter, and by that point the choice is usually between two member states rather than twelve.
The output of this exercise should be a written memo that records the rejected options and why. It is worth more than the recommendation itself: it is what you will hand to an investor, a board or a regulator asking why you chose this route, and it is what stops the question being reopened every quarter.
Next steps
See licensing for the regimes in detail, including capital thresholds, statutory review windows and refusal reasons, or send a description of the model and we will tell you which regimes are genuinely in scope.
Frequently asked questions
Can we start in one regime and move later?
You can, and it costs roughly what the first application cost, plus the disruption of migrating customers. It is a legitimate strategy when the first regime is genuinely a stepping stone, and an expensive accident when it was chosen carelessly.
How much weight should we give to state fees?
Very little at the selection stage. Fees are visible, comparable and therefore over-weighted. The costs that decide viability — key function holders, substance, audit, reporting — are none of those things.
Who should own this decision internally?
Whoever is accountable for the business plan, not the person managing the application. The choice constrains the commercial strategy for years, and delegating it to a process owner is how businesses end up with technically correct licences that do not fit what they sell.
Tell us what you are building
A short description of the product, the markets and the payment flows is enough for us to say what is required, in what order and at what cost.
Describe your matter
We reply within one business day with a scope, the deliverables and an indicative fee — not a brochure.