Revolut

IWhat it does

Revolut began in 2015 as a prepaid card and currency exchange app. Today it describes itself as a financial services group that gives customers one app for spending, saving, investing, borrowing and sending money¹. It had 68.3 million retail customers and 767,000 business customers at the end of 2025, across 39 countries¹. The group counts its customers as 69.1 million once business customers are added¹.

For an individual, the core is a current account with a debit card, with cheap foreign currency spending and transfers. On top of that sit paid subscription plans (the annual report names Premium, Metal and Ultra tiers), savings products, stock, commodity and crypto trading, personal loans, credit cards and, in a small way so far, mortgages in Lithuania¹. It also sells extras such as travel eSIMs, a points scheme called RevPoints, and a separate crypto exchange called Revolut X¹.

For companies, Revolut Business offers accounts, cards, payments, expense tools and merchant acquiring, which means helping a business accept card payments. The company says Business made up 16% of 2025 revenue, handled £277 billion of transaction volume, and passed $1 billion of annualised revenue¹. Annualised means a recent month or quarter multiplied up to a full year, so it is not the same as audited revenue.

The licensing picture matters because it decides what Revolut may do with customer money. It holds a banking licence from Lithuania (granted 2018) and operates as a licensed bank in 30 countries, according to the annual report’s highlights page¹. In the UK it exited a long restricted start-up phase in March 2026 (see the numbers section). It started full banking operations in Mexico in January 2026 and filed for a US national bank charter in March 2026¹. In September 2026 the US Office of the Comptroller of the Currency (OCC) gave conditional approval, which is a step short of being able to open¹⁷.

A note on scale of evidence. Almost everything above comes from the company’s own annual report, which was audited by BDO LLP and filed at Companies House on 3 April 2026¹². Audited does not mean independent of the company’s framing, so I separate audited numbers from the narrative around them.

IIHow it makes money

Revenue in 2025 was £4,516 million, up from £3,090 million in 2024¹. At roughly $1.34 to the pound (my assumption, set to match the company’s own conversions of about $6 billion revenue and $2.3 billion pre-tax profit⁷) that is about $6.1 billion. The company says 76% of turnover is fee based, which it calls unusual for a bank, and 21.6% is interest income¹.

Fee income is a collection of many small streams rather than one. Card payments, meaning the small cut Revolut earns when customers spend on its cards, were the largest at £1,004 million. Subscriptions were £708 million, wealth (trading and investing fees) £663 million, foreign exchange £606 million, the RevPoints rewards line £193 million, and other fees £255 million¹. The company says it now tracks 11 product lines that each exceeded £100 million of revenue in 2025¹.

Interest income was £974 million. It comes from three places: interest earned on cash held at central banks and other banks (£448 million), interest on treasury investments such as government bonds (£361 million), and interest on loans (£151 million)¹. A point worth noticing: interest earned on cash fell from £527 million to £448 million even though balances grew, which shows how sensitive this line is to interest rates¹. Interest income also grew only 23% against 46% for revenue¹, so it is becoming a smaller share of the whole.

Costs fall into three buckets. Cost of sales was £998 million, mostly fees paid to card networks and payment partners (£767 million), plus £81 million of credit losses and £83 million of interest expense¹. Administrative expenses were £1,804 million, of which staff costs were £922 million and advertising and marketing £529 million¹. Share-based pay inside staff costs was £193 million¹. Profit before tax was £1,713 million, tax £409 million, and net profit £1,305 million¹.

As a rough yardstick, I estimate revenue per average retail customer at about £75 in 2025. That uses the average of the opening and closing retail customer counts (52.5 million and 68.3 million) and includes business revenue, so it is only a guide¹. It shows how little each customer pays on average, which is why growth in customers and in products per customer both matter.

Revenue and profit, FY2025 vs FY2024 (reported, UK Companies House filing, £ millions)

IIIThe market and the competition

Revolut sits across several markets at once: everyday banking, cross-border payments, retail investing, consumer credit and business banking. That is a strength in that it can sell more to each customer, and a weakness in that it meets a specialist in every lane.

In its home region the company says 1 in 5 working-age adults in Europe now uses Revolut, and that nearly 1 in 3 newly opened bank accounts in six key markets is a Revolut account, citing survey data from the investor Andreessen Horowitz¹. That is a company-chosen statistic from a source with an interest in the industry, so I treat it as a hint of momentum, not proof of market share. The same report calls 2025 growth in France, Italy and Spain particularly strong¹.

The competition falls into four groups, in my judgement. First, traditional banks, which hold most salaries and mortgages and are slowly improving their apps. Second, other digital banks and challengers. Third, specialists in one function: Wise in international transfers, Robinhood in retail trading, SoFi in digital lending and membership, Nubank in Latin American digital banking. These are the listed companies I use as price yardsticks, and they show that a focused product can scale on its own. SoFi, for example, reported 17.1 million members, up 35%, and guided to about 40% revenue growth in 2026 and 27% in 2027^521181920. Fourth, large payments and technology firms that could add banking features.

Geography is the next test. Customer address data shows that 71% of 2025 fee income (£2,437 million of £3,429 million) came from Europe outside the UK, 24% from the UK (£840 million) and only 4% from the rest of the world (£152 million)¹. The company says its expansion outside Europe, in Mexico, the US, India and elsewhere, is its next frontier, and the CFO describes it that way⁷. Those markets are early and unproven.

In the United States, Revolut would be a late entrant against large banks and well-funded fintechs. The OCC’s conditional approval requires at least $95 million of starting capital and a tier 1 leverage ratio of at least 10% for the first three years after opening, and the bank still needs sign-off from the Federal Reserve and the FDIC¹⁷. The approval does not cover Revolut’s proposed retail foreign exchange business, which is one of the products it is best known for¹⁷. The company targets a 2027 launch¹⁷.

IVThe moat and how it could be disrupted

I rate the moat contestable. The advantages are real, but I cannot show they are durable, because most of the customer growth happened in a period of low switching friction and rising use of app-based finance.

What helps Revolut: scale and brand in Europe, a cost base that grew about 10% in average headcount while revenue grew 46%¹, 11 product lines above £100 million of revenue each to cross-sell¹, and a growing set of banking licences that let it hold deposits and lend. The company says customers hold more balances (+11%), transact more (+24%) and use paid plans more (+9%) year on year¹. The company also says more than 75% of customer support chats were resolved by a generative AI chatbot in 2025¹, which, if true at that scale, supports the low-cost claim. These are company statements, not audited figures.

What limits it: switching costs for a secondary spending account are low. A customer can use Revolut for travel or investing while keeping a salary at a traditional bank. The annual report shows rising balances per customer as a trend, but does not state how many customers use Revolut as their main account, so I cannot judge the depth of the relationship. The features, such as card controls, currency exchange and savings, can be copied by banks over time, and they have started to.

Regulation cuts both ways. A banking licence is hard to get, and it took Revolut roughly four years from application to full UK launch by one count (City AM) or about 20 months of restricted operation by another (Tech.eu), a gap I show rather than resolve⁹¹⁰. That protects incumbents and Revolut from newcomers, but it also means regulators have a say over product launches and growth speed. The annual report adds that failing to meet regulatory obligations can lead to fines, restrictions or conditions on its business¹.

Network effects are modest. Some features, such as paying a friend, improve with more users, but most of the value is individual. Data is a real asset in fraud detection and credit scoring, but every bank has data. So the moat is mostly scale, brand and execution speed, which are the type of advantages that can erode if the company slows down.

VThe numbers we can see

Because Revolut files group accounts at Companies House, we can see audited numbers, which is rare for a private company of this size. The 2025 accounts were approved by the board on 12 March 2026 and are listed on the register as filed on 3 April 2026¹². They are the strongest evidence on this page.

Profit has grown faster than revenue. Profit before tax rose 57% to £1,713 million, against revenue up 46%, which lifted the pre-tax margin to 38% from 35%¹. Net profit rose from £790 million to £1,305 million¹. The company calls 2025 its fifth profitable year in a row⁸. Revolut paid no dividends in 2025¹.

The balance sheet is a bank-style balance sheet. Total assets were £43.0 billion, of which cash and cash equivalents were £21.0 billion and treasury investments £17.6 billion, so about 90% of assets are cash or high-grade securities¹. Customer liabilities, essentially deposits and other money owed to customers, were £36.9 billion¹. The company also reports total customer balances of £50.2 billion, which includes money not on its own balance sheet; I do not have a breakdown of the difference¹. Loans to customers were £2.15 billion, up 120%, and the loan-to-deposit ratio is 6.2%¹. Expected credit loss coverage on loans was 4.1%¹. That is a very low-risk balance sheet today and a very different one from a lender that grows loans quickly.

Capital is where an owner should look. Total capital resources were £4.9 billion, all of the highest quality (called CET1)¹. Equity of £4.9 billion supports the US charter (minimum $95 million for the new bank) and the lending push¹⁷. The annual report states the group has capital and liquidity comfortably above regulatory requirements; it is a company statement, and I did not see the full regulator-by-regulator figures in the pages I reviewed.

Two figures from the register are worth a note. The Companies House record lists Nikolay Storonskiy as the only person with significant control, holding more than 25% but not more than 50% of the shares, with the right to appoint or remove directors³. It also shows a new set of articles of association adopted in September 2026, share allotments in 2026, and several filings for the company buying its own shares into treasury and re-issuing from treasury². I did not open those documents, so I cannot say what they contain.

UK banking: the Prudential Regulation Authority lifted the restrictions on the UK bank licence on 11 March 2026. Revolut had applied in 2021 and received a licence with restrictions in July 2024⁹. Press accounts differ on how long the process took, as noted in the moat section⁹¹⁰. The annual report says current accounts are being introduced gradually to UK customers, and the press reports 13 million UK customers¹¹⁰.

Selected audited or company-reported metrics (31 Dec 2025 unless stated)

VIWho runs it and who owns it

Nik Storonsky co-founded Revolut in 2015 and is group chief executive; Vlad Yatsenko co-founded it and is chief technology officer¹. Both are directors and have been since July 2020⁴. Martin Gilbert signs the annual report as chair; the other directors listed at Companies House are Caroline Britton, Michael Sherwood, John Sievwright, Dan Teodosiu and Ian Wilson, plus Siddhartha Jajodia, appointed 9 July 2026¹⁴. A press report calls Mr Jajodia the former US chief executive and names Cetin Duransoy as US CEO¹⁷.

On ownership, the register only says that Mr Storonsky holds between 25% and 50%³. A person familiar with the matter told the Irish Times he owns about 29% (press-reported, not filed)¹³. Other holders include venture and growth investors; I did not open a reliable and current cap table, so I will not state percentages for them. Shares in a private company are usually of different classes with different rights, and I have not seen those terms.

Three governance points deserve attention. First, new articles of association were adopted in September 2026². The Irish Times reported that an update, code-named Project Shasta, would raise the limit on the chief executive’s borrowing against his shares from $50 million to $250 million and remove the cap on the share of his holding he can pledge, and that Revolut called it a routine modernisation¹³. Second, the same report says Mr Storonsky is in talks over a new pay deal that raises his stake if the company reaches about $500 billion¹³. A separate report says an incentive package would give him roughly 40% of the company at a $200 billion valuation¹⁶. Those two press reports describe the pay arrangements differently, and I could not see the actual terms, so I treat them as a source of dilution for other holders, not a certainty. Third, the board audit tender replaced BDO with Ernst & Young for 2026, following a formal tender¹.

The chief executive’s stake is strong alignment, but it also concentrates control. Together with the right to appoint directors³, an outside investor has limited say. The company has run share sales roughly every one to two years, and Mr Storonsky says there may be more before any IPO¹⁴. Employees and early investors rely on those sales for liquidity, since there is no dividend and no public market.

VIIValuation: is the last price supported?

Equity value in US$ billions. Not a share price: a private company has none. Latest mark: secondary sale (employee liquidity, no new money), press-reported.

What the last valuation assumes

At $115 billion, and using my base multiple (20 times earnings), discount rate (12%) and dilution (2% a year), the price needs 2030 net profit of about $10 billion. That is about 5.8 times 2025 net profit of about $1.7 billion (my conversion of £1,305 million¹). At a 30% net margin, that means revenue of about $34 billion, which is roughly 5.5 times 2025 revenue and a growth rate near 41% a year for five years. If the market were still paying SoFi’s 28 times in 2030, the required profit falls to about $7 billion. Either way the price assumes the company keeps growing fast, keeps its margin, avoids a regulatory or credit setback, and is still valued like a premium growth company at the end.

Bear, base and bull

• Bear. Revenue growth slows from 28% in 2026 to 10% in 2030, reaching about $13.2 billion in 2030. Net margin falls to 22% from 29% in 2025 on rate cuts and credit costs, so 2030 net profit is about $2.9 billion. I apply 12 times earnings, below Nubank’s 14.1x¹⁸, discount at 14% a year, and allow 1.5% a year of dilution. All of this is my estimate. Result: about $19 billion.
• Base. Revenue grows 38%, 32%, 26%, 21% and 17% a year to 2030, reaching about $19.7 billion (£14.7 billion), from a 2025 base of £4,516 million¹. Net margin is 30%, close to the 2025 level of 29%¹, giving 2030 net profit of about $5.9 billion. I apply 20 times earnings, near Wise at 19.4x²⁰ and between Nubank at 14.1x¹⁸ and SoFi at 28.4x¹⁹. I discount at 12% a year and allow 2% a year of dilution for staff awards and the founder incentive. Result: about $67 billion. FX is $1.34 to the pound (my assumption).
• Bull. Revenue grows 45%, 40%, 34%, 28% and 22%, reaching about $25.7 billion. Net margin rises to 34% as fee income scales, so 2030 net profit is about $8.7 billion. I apply 28 times earnings, like SoFi’s 28.4x¹⁹, discount at 11%, and allow a heavier 3% a year dilution because the founder incentive reportedly pays out in this world¹⁶¹³. Result: about $139 billion.

Comparables used

Where I come down

The reported $115 billion is above my base of $67 billion, by roughly 1.7 times, and sits between my base and bull cases ($67 billion and $139 billion). It also exceeds the ’margin of safety’ price of $50 billion. To be right, an owner would have to believe that revenue can grow about five-fold in five years, that margins hold near 30% while the lending book and US entry add cost, and that investors still pay a premium multiple for the shares in 2030. The market may be right even though it looks high. Revolut grew revenue 46% and profit before tax 57% in 2025¹, the balance sheet carries little credit risk so far¹, and buyers in a secondary sale may be pricing an IPO at $150 to $200 billion, which the company has reportedly discussed with investors¹⁵. Also, a private sale price reflects limited supply of shares, which can sit above what a broad public market would pay. The Forge Global mark of about $121 billion in October suggests demand has not faded¹⁶. I would not rely on one forecast: my range is wide, from $19 billion to $139 billion.

What I cannot judge

I cannot see the terms of the founder incentive plan, so I do not know how much value could shift from other holders; I used a heavier dilution rate in the bull case to reflect this. I do not have the share classes, the full cap table, or the number of shares that changed hands in the July 2026 sale, which Revolut confirmed was under way without confirming the valuation¹¹; I found no announcement that it completed. The price per share ($2,017) is press-reported by Bloomberg and the Wall Street Journal via secondary reports, and I did not see their original articles¹¹¹². I cannot judge credit quality through a recession, the pace of rate cuts, or how regulators in the UK, EU and US treat the company over the coming years. Comparable-company data comes from aggregator pages whose dates and figures disagree with each other (for example, Nubank is shown at $56 billion on the page I opened¹⁸, while a search snippet I did not open gave a different figure), so the multiples are rough. Pre- and post-money are the same here because no new shares were sold for cash¹². Finally, the model is mine and is built on a small number of inputs, so a wide range is the honest answer.

VIIIThe IPO

Reported to be preparing an IPO. Revolut has not filed publicly or, as far as I can find, confidentially. The CEO said in April 2026 that a listing is about two years away, so not before 2028¹⁴. The Financial Times reported that Revolut told investors it targets an IPO valuation of up to $200 billion and discussed a range of $150 billion to $200 billion, and a source said no formal valuation has been decided¹⁵. A separate report says he reportedly prefers a US listing¹⁶. No exchange, ticker, underwriting banks or offering size appear in the sources I opened. A search of the SEC’s company list finds only Revolut Securities Inc., a broker-dealer, and I saw no registration statement from the group⁶. The CEO also said further secondary sales may happen before any IPO¹⁴.

What to watch

• A public or confidential IPO filing, and the choice of exchange (the CEO is reported to prefer the US)¹⁶
• Whether the July 2026 secondary sale is announced as complete, with size and price¹¹
• The 2026 annual report, due at Companies House in 2027, for revenue, margin and capital¹
• Final OCC, Federal Reserve and FDIC approvals for Revolut Bank US and its 2027 launch¹⁷
• Loan growth and credit losses as the UK bank begins lending¹⁹
• The terms of the founder incentive and the share-pledge changes in the new articles²¹³
• Listed fintech multiples (SoFi, Wise, Nubank, Robinhood) as the reference point for any IPO price^19201821

IXWhat could go wrong

The most likely way to destroy value is not a collapse in customers. It is a combination of a regulator slowing growth, interest rates falling while the company is still dependent on them, and a credit problem once lending grows. Each is reasonable on its own. I list them in the order I think matters.

Regulation and conduct. The UK licence took years and ran well beyond the usual restricted period of about 12 months, according to Tech.eu⁹. Banks that scale quickly face tighter scrutiny on anti-money-laundering, product approval and customer protection. The annual report names fines, public censure, suspensions and restrictions as possible outcomes of non-compliance¹. The company’s own auditor flagged a heavy reliance on IT systems in its audit work¹, which is a reminder that an outage or fraud wave is a regulatory event as well as a customer one.

Interest rates. Interest income was 21.6% of revenue¹, and interest on cash balances fell in 2025 even though deposits grew¹. If central banks cut rates further, this line shrinks unless balances grow faster than rates fall. The company has been shifting to fees (76% of revenue), which helps, but fees also depend on market activity, for example wealth fees depend on trading volume and crypto prices.

Credit. Loans grew 120% to £2.15 billion¹ and credit losses on lending doubled to £61 million¹. Loans are still only 6.2% of deposits¹. A UK bank licence, a US charter and mortgages push toward more lending. Lending to people who chose a cheap app, not a branch, has not been tested through a hard recession at this scale.

Valuation and liquidity. A secondary price is set by who is willing to buy a small number of shares. The buyers were described only as new and existing investors, and the size was not disclosed¹¹. If listed fintech multiples fall, or if the company cannot run another sale, the price can reset quickly. Forge Global, a private-share marketplace, marked the stock at $2,125 on 15 October 2026, about $121 billion, but that is an indicator, not a completed round¹⁶.

Key person and governance. The chief executive is both the founder and the largest holder, and he is negotiating a pay deal linked to value targets and has asked to raise the limit on borrowing against his shares¹³. This is not unusual for founder-led companies, but it means outsiders depend on one person’s judgement and incentives.

Expansion cost. The US bank requires at least $95 million of starting capital and a 10% leverage ratio for three years¹⁷. Mexico, India and other markets need local capital and years of customer acquisition. Marketing spend rose 47% in 2025¹. If new markets take longer to pay back than Europe did, margins fall.