Revolut’s 2025 results suggest that high growth and bank-grade profitability can coexist. But operating as a licensed bank across dozens of countries creates a different challenge from distributing one elegant app: deposits, lending, capital, compliance and local service obligations do not scale at the speed of software.
The hidden tension in a global banking super-app
Revolut’s defining product advantage has been low-friction access to cross-border payments, subscriptions, investments and everyday money management. Its global banking ambitions now test whether this distribution model can survive the economics of traditional banking. The firm reported $6.0 billion of revenue in 2025, a 38% pre-tax margin, 68.3 million retail users and $67.5 billion of customer balances [1]. These figures create a more serious challenge to legacy banks than user-count comparisons alone. They also show a business whose financial exposures are changing.
Consider the distinction between permission to offer an app and permission to accept insured deposits. The UK Prudential Regulation Authority cleared Revolut Bank UK Ltd to exit its mobilisation phase on March 11, 2026, and eligible customers were to migrate in stages [2]. APRA granted a banking licence in Australia on July 21, 2026 [3]. The US OCC registry records a September 2, 2026 approval of a US national-bank-charter application [4]. Each step is significant, but the customer experience still depends on the specific entity, operational readiness, applicable conditions and product availability. A public approval entry does not alone demonstrate that every US account can already use a newly opened national bank.
The claim that “Revolut is a bank” is therefore less informative than identifying which legal subsidiary holds the balance, which supervisor oversees it, and what type of claim the customer has. digital-bank fees, features and deposit protections explains why the same brand may sit behind different customer protections in different markets.
Where its fintech economics are strongest
| Revenue line | Revolut 2025 | Main driver | Main risk |
| Card payments | $1.3bn | Frequency × take rate | Network / fraud |
| Subscriptions | $936m | Paid adoption | Downgrades |
| Wealth | $876m | Assets/activity | Markets / conduct |
| FX | $800m | Transfer volume/margin | Competition |
| Interest income | $1.3bn | Balances/yield | Rates / funding |
Table 1. Reported 2025 selected streams; company cautions they do not reconcile to total revenue.
In 2025, Revolut reported $1.3 billion from card payments, $936 million from subscriptions, $876 million from wealth, $800 million from FX, and $1.3 billion of interest income [1]. These are not an exhaustive additive bridge to total group revenue: the company’s published categories and translated figures are illustrative. They do, however, map a portfolio that combines high-frequency payments, premium-product monetization, financial-market participation and balance-sheet income. That mix can stabilize the company when one channel weakens, but it also creates different risk and regulatory profiles.
Card activity benefits from many small, recurring consumer decisions and merchants accepting network credentials. Subscriptions bring more recurring payments but can be vulnerable when benefits are copied or customers downgrade. FX revenue depends on volume, customer mix and pricing, not just a headline “zero fee.” Wealth income is market-sensitive and involves suitability, custody or conduct obligations depending on the product. Interest income can benefit when balances are large and interest rates are favorable, but it is far from a pure software subscription.
Revolut Business contributed 16% of group income in 2025, the company said, while business customers grew 33% to 767,000 [1]. Corporate relationships have higher potential lifetime value, but they need reliability, controls, invoicing, reconciliation and support. The cost of a payment failure or compliance hold for a merchant can be much larger than the monthly fee earned from its account.
why international transfer costs exceed headline bank fees show why “faster and cheaper” has more moving parts than a displayed transfer price. True cost depends on funding, exchange-rate margin, payout rail, intermediaries, settlement time, refunds, customer support and who absorbs loss when something breaks. A fintech can simplify the interface while still bearing these underlying banking costs.
Why a software app cannot simply multiply banking licences
The marginal cost of showing the same button to another million users may be low. The marginal cost of opening in another banking jurisdiction can be high. Each licence introduces entity governance, management, reporting, capital or liquidity requirements, financial-crime monitoring, regulatory engagement, access controls, complaints handling and consumer-disclosure obligations. Some rules can be shared through group infrastructure, but accountability and legal requirements cannot be dissolved into a single central API.
This resembles a franchise constrained by local regulatory balance sheets, rather than a global social-media product. The distinction is especially important where lending grows: loans are assets that earn interest but absorb capital and can default. Revolut’s outstanding customer lending portfolio rose 120% to $2.9 billion in 2025, according to the group [1]. That is still modest compared with reported customer balances, but a doubling credit book may change the mix of required risk expertise faster than a user-count chart suggests.
The UK launch also illustrates a second operational cost. Revolut’s own notice said eligible existing customers would receive advance notice and be migrated in batches; some new sign-ups could initially be placed with the e-money entity rather than the bank [2]. Running two legal arrangements during transition creates customer-education, compliance, communications and operational complexity. A depositor checking their statement and deposit-protection eligibility may care more about that distinction than about the quality of the user interface.
A customer-balance model exposes the growth requirement
Figure 1. Group customer balances divided by year-end retail users: an imperfect intensity proxy, not reported average retail deposits.
Revolut’s total reported customer balances rose from approximately $38 billion in 2024 to $67.5 billion in 2025, while year-end retail customers increased from 52.5 million to 68.3 million [1]. Dividing total balances by year-end retail users produces $724 and $988, respectively—an increase of about 36.5%. This is a deliberately rough group-wide “balance intensity” indicator, not average individual deposits, official active-user balances or customer lifetime value. It includes balances attributable to business relationships while the denominator excludes business accounts. Its analytical use is to show how growth can come from more valuable financial relationships, not only new registrations.
If retail user counts grow faster than balance intensity, the company must monetize users with payments, wealth or subscriptions. If customer balances grow faster, it may gain interest income and loyalty but also incurs the duties attached to safeguarding or accepting those funds. An additional $10 billion of interest-sensitive assets earning one percentage point more would mathematically generate $100 million in gross annual interest before liability repricing, hedging, taxes, liquidity buffers and operating costs. This is not a prediction of Revolut’s yield or its legally investable balances. It illustrates why a global bank’s fortunes may increasingly respond to interest rates and deposit behavior.
A reverse scenario makes the risk clearer. Revolut reported roughly $1.3 billion of interest income in 2025 [1]. A hypothetical 20% decline in that revenue line, with all other income and expenses unchanged, removes $260 million, equivalent to about 11% of its reported $2.3 billion pre-tax profit. At a 40% shock, the mechanical reduction is $520 million, about 23% of pre-tax profit. Actual outcomes would be affected by liability costs, market rates, balance growth, asset mix, hedging and management decisions; the model is a vulnerability test, not earnings guidance.
Figure 2. Hypothetical interest-income decline, with 2025 pre-tax profit and every other variable frozen; not guidance.
Can credit deepen the relationship without damaging returns?
Banking access lets a provider extend credit to existing customers, but the unit economics of lending depend on loan pricing, credit losses, funding, underwriting cost and capital. A consumer loan at 12% nominal interest is not a 12% profit stream. For a $10,000 illustrative loan, 12% of interest income is $1,200 a year. If expected credit losses equal 4% ($400), funding costs 3% ($300), servicing and fraud expense 1.5% ($150) and incremental capital/compliance expense 1% ($100), the simplified annual contribution is $250, or 2.5%. A two-percentage-point deterioration in expected losses cuts that contribution to $50. These percentages are illustrative, not disclosed Revolut lending metrics.
Compare that $250 result with a payments customer who generates fee and FX revenues without an unsecured lending book: their profit may be lower, but their credit-risk and capital demands may also be lower. This is why “more bank products” does not automatically mean better margins. The goal is return on risk-adjusted capital. The consumer’s perspective is equally important: competitive credit can be helpful, yet a seamless approval flow could encourage over-borrowing unless affordability, customer understanding and complaints processes keep pace.
The firm’s customer-protection investments are not optional brand expenses. Revolut reported using AI across its customer-protection platform, increasing the volume of potential fraud cases reviewed daily by tenfold [1]. The economic question is whether automation reduces fraud losses and staffing per incident without creating more false positives, frozen accounts or failed appeals. Efficiency must be measured after the cost of wrongly blocking legitimate customers.
The three-layer cost of a global expansion programme
| Economic layer | Operational requirement | Investor KPI |
| Market entry | Licensing, build, onboarding | Launch spend / milestones |
| Continuing overhead | Compliance, audit, service | Contribution per active user |
| Balance-sheet capital | Liquidity, lending capital | Risk-adjusted return |
Table 2. Analytical framework, not Revolut country-level budget disclosures.
Expansion has at least three layers: entry cost, ongoing local control costs, and capital tied up to support bank products. The first includes licensing, integration and launch teams. The second includes local financial-crime specialists, customer-service coverage, security, audit, legal and recurring supervisory interactions. The third includes reserves, liquidity and credit capital that cannot simply be paid out to shareholders.
Figure 3. Hypothetical break-even = $15m annual fixed expense divided by annual contribution per active customer; excludes entry cost and capital.
Consider a completely hypothetical new country requiring $25 million upfront, $15 million recurring annual fixed expense, and producing annual contribution revenue net of variable servicing costs of $10, $25 or $50 per active customer. Excluding capital and tax, its annual break-even active-user count is 1.5 million, 600,000 or 300,000 respectively. The $25 million initial outlay remains to be earned back. If a bank requires another $100 million of regulatory capital, that sum is not automatically an expense, but it has an opportunity cost and must earn an acceptable return. These models are deliberately not estimates of the actual cost to enter any specific market.
Scaling to dozens of new territories may create a portfolio where successful markets subsidize expensive markets for years. A public-market investor should ask for profitable mature cohorts by country, the cost to bring new charters online and the share of revenues generated without requiring substantial credit capital. A bank with the strongest app is not necessarily the bank with the best incremental return on deployed capital.
The risk of losing the advantage it was built on
The fintech advantage can erode through four mechanisms. First, a proliferating local legal architecture makes seemingly universal products differ by residency, undermining the simplicity of the interface. Second, heavy credit growth shifts management focus from product experimentation to defaults, recoveries and balance-sheet quality. Third, restrictions, account reviews and fraud controls may scale less smoothly than transaction throughput. Fourth, a larger group may need slower change-management processes, including independent testing before releasing features that touch money or customer rights.
These tensions do not mean regulators are the enemy of innovation. Deposit protection, safeguarding and financial-crime controls can make customers more comfortable using an app for salary and savings, enabling more economically valuable relationships. The proper trade is not frictionlessness against regulation; it is consistent customer trust achieved at reasonable operating cost. An app that opens quickly but routinely blocks ordinary transfers has not eliminated friction—it has shifted it from onboarding to the moment a customer urgently needs money.
card refunds and the underlying payment processing chain provide a concrete example: a polished interface cannot remove the operational sequence involving merchant, acquirer, scheme and issuer. A global finance platform can differentiate itself by making exceptions traceable, not by pretending third-party systems do not exist.
Three outcomes that could settle the argument
The optimistic scenario is that Revolut retains strong organic acquisition, expands active salary and business banking, converts more users to premium plans and builds sensible credit products. Its technology platform spreads central development costs over a rising revenue base, while strong local controls reduce fraud and regulatory risk. In that scenario, banking charters strengthen fintech economics rather than replace them.
The middle scenario is slower but still attractive: product growth continues, yet staff, cyber resilience and country-specific compliance expenses absorb much of operating leverage. Profit still grows, but margin expansion stalls. The pessimistic scenario combines deposit repricing pressure, rising credit losses, expensive international market entry and reliability problems. New charters could expand potential revenue while reducing returns on equity. The scenarios are qualitative, not forecasts.
What evidence would falsify the optimistic thesis? A sustained fall in net new active customers, deteriorating revenue per retained customer, rising complaint or fraud losses, increasing cost to serve per transaction, weak return on new-market capital, and poor performance as the lending book seasons. What would strengthen it? Improving active relationships, steady or rising customer balances with controlled funding cost, repeatable profitable expansion and measured fraud/complaints quality—not merely a billion-payment volume headline.
What customers and investors should watch
Customers should confirm the regulated entity named in their account terms, deposit scheme coverage where relevant, the difference between bank deposits and investments, applicable cross-border fees, complaint escalation and the availability of a second practical payment option. Investors should look for a return-on-capital and credit-loss bridge, rather than compare Revolut’s price-to-sales multiple with a traditional bank’s price-to-book multiple without adjustment.
Revolut’s five-year £10 billion investment commitment is an ambition announced by management, not proof that this spending has all been incurred [1]. A public company would need to show how it balances growth investment with group profitability and how much of its balance sheet is restricted inside regulated entities. Investors would also want to examine whether reported customer balances are bank deposits or other legally different liabilities, what fees are charged on each product, and which assets can actually support lending.
Bottom line
Revolut has already demonstrated that a multi-product finance app can reach substantial scale and profitability. Becoming a global bank could reinforce that advantage by moving more customers’ primary financial relationships onto the platform. The risk is not that regulation necessarily ruins fintech; it is that credit capital, local controls and exception handling grow faster than contribution per customer. The most revealing metrics are balance intensity, credit losses, local operating costs and risk-adjusted profitability. Until those can be tracked consistently across countries, “global bank” remains a strategic goal rather than proof of an endlessly scalable financial model.
Methodology
Reporting cut-off October 18, 2026. Company metrics are reported 2025 full-year results and 2024 comparatives, with USD figures rounded as the company presents them. Balance intensity divides reported group customer balances by retail user totals and is not an individual deposit or true ARPU statistic. Interest-income shock holds all other income and expense constant, which is intentionally unrealistic and therefore only a sensitivity. Loan example and country-entry economics use purely invented model assumptions and are not Revolut guidance. Charter milestones are distinguished from actual account rollout, customer migration and deposit-protection applicability. Local consumer protections, banking permissions and business features are jurisdiction- and entity-specific.
Sources
[1] Revolut, March 24, 2026 — full-year 2025 results — Open source
[2] Revolut, March 11, 2026 — UK bank launch and migration — Open source
[3] APRA, July 21, 2026 — Revolut’s Australian ADI licence — Open source
[4] OCC — US charter application and September 2026 approval entry — Open source
[5] Revolut, March 5, 2026 — US charter filing — Open source
[6] Revolut — annual report 2025 landing page — Open source
[7] Revolut UK help — phased bank transition — Open source
[8] Revolut UK help — business eligibility for bank migration — Open source
[9] Revolut UK help — changing Revolut Pro provider after migration — Open source
[10] Companies House — Revolut Ltd filing history — Open source
[11] Companies House — Revolut Bank UK Ltd accounts and filings — Open source
[12] Reuters, October 9, 2026 — ambitions beyond banking and AI models — Open source
[13] Bank of England, July 2026 — frontier AI and financial-stability vulnerabilities — Open source
Johan Shamshad is a financial markets writer at Dave Finances covering cryptocurrencies, trading platforms, brokers, fintech, financial regulation, and developments across global markets. He previously worked at Gulf News, adding newsroom experience to his coverage of fast-moving financial and digital-asset markets.
His work focuses on identifying market-moving events, company developments, regulatory changes, product launches, and shifts in trading and financial infrastructure.
Johan contributes news and analysis designed to help readers understand not only what happened, but why a development matters and how it may affect the wider financial landscape. You can reach out to him via his social media accounts:
Linkedin: https://www.linkedin.com/in/johan-shamshad-742851262/
X: https://x.com/Yasmine_FX
Investing: https://www.investing.com/members/contributors/279781574