Branchless banking removes a visible cost center, but it does not create a business model. The economics come from interchange, deposits, lending, subscriptions, wealth, foreign exchange and paid liquidity—and the mix differs dramatically between Chime, Revolut, Monzo and Nu.
| Metric | Latest disclosed figure | Why it matters |
| Chime Q2 2026 | $670m revenue; 64% payments, 36% platform-related | A payments-led model that is diversifying beyond interchange. |
| Revolut FY2025 | £4.5bn revenue; card payments only 22.2% of turnover | A multi-product fee model, not simply a card business. |
| Monzo FY2026 | £1.7bn revenue; £25.7bn deposits; £73bn card spend | A licensed-bank model monetizing balances as well as activity. |
| Nu Q2 2026 | $5.88bn managerial revenue; $17.1 ARPAC vs $1.0 cost to serve | A credit-and-float engine built on unusually low servicing costs. |
The Core Question: What Replaces the Branch?
The easiest way to misunderstand a neobank is to treat “no branches” as the answer to how it makes money. Branches are primarily a distribution and service channel. Removing them can lower fixed costs and make geographic expansion faster, but revenue still has to come from the same economic raw materials that power traditional banking: payments, deposits, credit, fees and financial intermediation.
What changes is the operating architecture. A branchless platform can acquire customers through an app, automate onboarding, centralize servicing, use transaction data for underwriting and sell multiple products to the same customer without opening a new location. But it also inherits a different set of costs: cloud infrastructure, card-network fees, bank-partner economics, fraud losses, compliance, customer support, incentives and digital marketing.
That distinction matters because the label “neobank” covers several structurally different businesses. Chime is an asset-light technology company that distributes banking products through partner banks. Monzo is a licensed bank with a deposit-funded balance sheet. Revolut operates a multi-entity banking and financial-services group. Nu is a full digital bank whose economics are increasingly dominated by credit and deposit spread. They all lack large branch networks, but they do not make money in the same way.
Four Branchless Models, Four Different Profit Engines
| Company | Operating model | Latest monetization evidence | Primary economic engine |
| Chime | U.S. fintech using FDIC-insured partner banks | Q2 2026: $430m payments revenue and $240m platform-related revenue | Interchange plus paid liquidity/transfer products |
| Revolut | Multi-product financial group with bank entities | FY2025: 76% of turnover fee-based; 21.6% interest income | Card, subscriptions, wealth, FX and interest |
| Monzo | UK licensed digital bank | FY2026: £25.7bn deposits and £1.7bn revenue | Deposit balances, borrowing, payments and subscriptions |
| Nu | Latin American digital bank | Q2 2026 managerial revenue: 61% credit, 25% float, 14% fees | Credit spread plus deposit/float income |
1. Interchange: The Simplest Neobank Revenue Engine
When a customer pays with a debit or credit card, the merchant’s acquiring side ultimately sends an interchange fee to the card issuer. For a fintech such as Chime, the partner bank is the legal issuer, collects the interchange and passes an agreed amount to Chime. Chime recognizes its payments revenue based on those interchange fees.
In Q2 2026, Chime generated $430 million of payments revenue on $38.0 billion of purchase volume. That is equivalent to about 1.13% of purchase volume. This is not a universal interchange rate—its mix includes debit and secured-credit activity with different economics—but it shows why everyday card usage can support a large revenue stream even when the customer pays no monthly account fee.
The vulnerability is regulation. U.S. Regulation II generally limits covered debit-card interchange, while smaller issuers can qualify for an exemption. Federal Reserve data for 2024 show average interchange of $0.51 per exempt transaction versus $0.23 per covered transaction. On 100 million transactions, that 28-cent average gap would equal $28 million of issuer-side economics. Chime explicitly warns that its model depends in part on its bank partners maintaining the small-issuer exemption.
Figure 1. Federal Reserve 2024 average debit interchange. The gap illustrates why issuer structure can materially change neobank economics.
The same model is much less generous in some other markets. The UK caps domestic consumer interchange at 0.2% for debit and 0.3% for credit. That helps explain why European neobanks have had to diversify faster into subscriptions, FX, wealth, lending and interest income instead of relying on card spending alone.
2. Deposits Turn a Banking App Into an Interest-Earning Balance Sheet
A branchless bank becomes economically more powerful once customers use it as a primary account and leave meaningful balances behind. Deposits can fund loans, sit in central-bank reserves or be invested in high-quality liquid assets. The bank earns income on those assets and pays customers some portion back as deposit interest. The spread is one of the oldest banking businesses; the innovation is doing it with a much lower servicing footprint.
Monzo’s FY2026 numbers illustrate the scale effect. Deposits rose 55% to £25.7 billion, while revenue rose 39% to £1.7 billion and card spend reached £73.0 billion. Monzo explicitly divides revenue into current-account balances, borrowing, payments, wealth, and subscriptions/other fees. The app may look like a payments interface, but the economics increasingly resemble a diversified bank balance sheet.
Nu provides an even cleaner decomposition. Its Q2 2026 managerial P&L reported $5.88 billion of revenue: $3.60 billion of credit income, $1.45 billion of float income and $816 million of fee income. In percentage terms, that is roughly 61% credit, 25% float and 14% fees. A customer does not need to swipe a card for Nu to earn money; simply holding deposits or carrying a loan can create revenue.
3. Lending Is Often the Most Powerful—and Most Dangerous—Step
Interchange monetizes spending. Lending monetizes balance-sheet risk. Credit cards, personal loans, overdrafts and salary-linked advances can generate much more revenue per customer than a debit card, but they also introduce funding costs, credit losses and capital requirements.
Nu’s Q2 2026 figures show the trade-off clearly. Its $3.60 billion of managerial credit income was paired with $1.69 billion of cost of credit and $1.37 billion of funding cost across the broader business. The branchless advantage therefore does not abolish banking risk; it changes the fixed-cost base around that risk.
Chime is moving in the same direction from a lighter starting point. Its platform-related revenue grew 48% year over year in Q2 2026, helped by MyPay, Instant Loans and outbound instant transfers. The company’s payments revenue still dominates, but its newer products show the natural evolution of a payments-led neobank: first win the transaction account, then monetize liquidity needs and financial relationships around it.
Figure 2. Chime’s disclosed mix shows platform-related revenue gaining share as the product set expands beyond card interchange.
4. Subscriptions, FX, Wealth and Convenience Fees Fill the Gaps
A neobank with limited lending appetite can still deepen monetization by charging for premium functionality rather than basic account access. Revolut is the clearest example. In 2025, card payments represented 22.2% of turnover, interest income 21.6%, subscriptions 15.7%, wealth 14.7% and FX 13.4%. The remaining 12.4% came from other activities.
Figure 3. Revolut FY2025 revenue mix. No single product line dominates, reducing dependence on one fee pool or interest-rate environment.
That diversification is strategically important. Interchange can be capped. Interest income falls when policy rates decline. Trading and crypto revenue can be cyclical. Subscriptions, business accounts, wealth services and foreign exchange give the platform additional ways to monetize the same acquired customer.
Monzo is using the same playbook from a banking base. Its 2026 report says more than 1.6 million customers had subscriptions, while its revenue categories also include wealth and payments. The retail implication is that “free banking” usually means the entry account is free—not that the relationship is economically free. Revenue is collected elsewhere in the stack.
5. The Real Branchless Advantage Is Unit Economics
The strongest case for branchless banking is not that it eliminates cost, but that the marginal cost of serving one more active customer can be unusually low once the technology platform is built.
Nu reported monthly average revenue per active customer of $17.1 in Q2 2026 and monthly average cost to serve of $1.0. That is a 17.1-to-1 ratio before funding costs, credit losses, taxes and corporate overhead. It should not be read as a 94% margin. It does show why a digital bank can add millions of customers without needing a proportional increase in physical locations or frontline staff.
Figure 4. Nu Q2 2026 unit economics. ARPAC is not profit, but the gap to service cost demonstrates the operating leverage of a scaled branchless platform.
Nu’s Q2 2026 managerial figures reinforce the point: 139 million customers generated $5.88 billion of quarterly revenue while customer support and operations expense was $226 million. AI and automation are beginning to widen that gap further; Nu says AI agents already handle more than 60% of customer-support conversations in Brazil at or above human parity.
No Branches Does Not Mean No Expensive Infrastructure
The branchless narrative can become misleading when it implies that digital banks operate with almost no overhead. Chime’s 2025 financials are a useful counterexample. The company reported $2.19 billion of revenue, but sales and marketing expense was $635 million, member support and operations was $458 million, and transaction and risk losses were $407 million. Those three lines alone totaled about $1.50 billion, or 68.6% of annual revenue, before technology, administration and other expenses.
The cost center has moved. Instead of leases, teller payroll and local branch operations, the digital model can spend heavily on customer acquisition, fraud reimbursement, card processing, bank partners, compliance, cloud infrastructure and 24/7 support. This is why scale matters so much: the technology advantage becomes economically meaningful only if revenue per active customer rises faster than these variable and semi-fixed costs.
Original Stress Test: What If Interchange Economics Weaken?
Chime’s Q2 2026 payments revenue of $430 million on $38.0 billion of purchase volume implies reported payments-revenue intensity of roughly 113 basis points. That makes it possible to illustrate how sensitive a payments-led model could be to a lower blended take, holding purchase volume constant.
| Illustrative decline in revenue intensity | Quarterly revenue impact | Share of Q2 2026 total revenue |
| 10 basis points | $38.0m | 5.7% |
| 20 basis points | $76.1m | 11.4% |
| 30 basis points | $114.1m | 17.0% |
| 50 basis points | $190.2m | 28.4% |
This is a sensitivity model, not a forecast. Actual interchange economics vary by card type, network, merchant category, partner agreement and regulation. But it demonstrates why payments-led neobanks keep adding lending, instant-transfer and subscription revenue: diversification can turn a regulatory or network-pricing shock from an existential problem into a manageable one.
A Better Way to Compare Neobanks
| Question | Why it matters | What to look for |
| Who owns the deposits? | Determines who earns float and carries liquidity obligations. | Partner bank vs licensed bank; deposit growth; funding cost. |
| Who takes credit risk? | Lending can expand revenue but introduces losses and capital needs. | Loan book, NIM, cost of credit, delinquency. |
| How dependent is revenue on interchange? | Payment economics are exposed to regulation and network pricing. | Payments share, purchase volume, issuer structure. |
| How expensive is acquisition? | A branchless bank can simply swap rent for digital marketing. | Marketing expense, organic referrals, payback period. |
| What does an active customer generate? | Shows whether scale is translating into monetization. | ARPAC/ARPU, primary-bank share, product adoption. |
| How much does servicing cost? | Tests whether digital operations actually create leverage. | Support expense, cost-to-serve, automation. |
What Retail Customers Should Understand
The economics explain why many neobanks can offer no monthly account fee, early direct deposit, free budgeting tools or attractive savings features. A customer who appears to be paying nothing can still be valuable through card interchange, deposit spread, cross-sold credit or a premium upgrade.
That does not make the model inherently exploitative. It does mean customers should identify which behavior the platform is trying to monetize. A free checking account may be subsidized by merchant-paid interchange. A high-yield savings account may be designed to attract low-cost funding. An instant-pay feature can become fee revenue. A credit product can be far more profitable than the deposit account that originally acquired the customer.
The legal structure also matters. Chime users are accessing deposit products provided by partner banks, while Monzo and Nu operate banking entities themselves. The user experience can look almost identical, but the entity holding deposits, extending credit and carrying regulatory capital can be different.
What Would Break the Branchless Thesis?
The branchless model is strongest when digital acquisition stays efficient, customers consolidate more of their financial lives onto the platform, fraud remains controlled and servicing costs scale slowly. The thesis weakens if customer acquisition becomes permanently expensive, interchange caps compress payment economics, deposit competition forces funding costs sharply higher, credit losses outrun pricing, or customer-service complexity requires a much larger human operating layer.
There is also a strategic paradox. As successful neobanks mature, they begin to look more like the banks they were supposed to disrupt: they add deposits, credit cards, loans, wealth, business banking and regulatory capital. The enduring advantage is therefore unlikely to be “we do not have branches.” It is whether the company can run a full banking relationship with materially better software, lower servicing cost and higher product velocity.
Bottom Line
Neobanks make money without branches because branches were never the source of banking revenue. Payments, deposits, lending and financial services were. The digital model changes how cheaply those products can be distributed and serviced.
Chime shows the payments-led version: 64% of Q2 2026 revenue still came from payments, but platform-related revenue is growing faster. Revolut shows the diversified-fee version, where card payments are only about one-fifth of turnover. Monzo shows the deposit-led bank becoming a multi-product platform. Nu shows the end state most clearly: enormous credit and float income layered on top of a reported $1 monthly cost to serve an active customer.
For investors, the most important metric is therefore not branch count. It is the spread between what each active customer generates and what it costs to acquire, fund, serve and protect that relationship. The best branchless banks do not merely remove real estate from the model. They turn a low-cost digital account into a high-frequency distribution channel for the entire financial stack.
Methodology
Figures are based on the latest public disclosures available as of October 8, 2026. Company reporting periods differ: Chime data use Q2 2026 and FY2025 disclosures; Revolut uses FY2025; Monzo uses FY2026; Nu uses Q2 2026. Revenue categories are not directly comparable across companies because accounting standards, product mixes and legal structures differ.
Derived calculations: Chime Q2 payments share = $430m / $670m; reported payments-revenue intensity = $430m / $38.034bn purchase volume; Chime FY2025 payments share = 49% debit-interchange revenue + 20% credit-interchange revenue; Federal Reserve interchange ratio = $0.51 / $0.23; 100m-transaction revenue gap = ($0.51 – $0.23) × 100m; Nu ARPAC-to-cost-to-serve ratio = $17.1 / $1.0; Chime 2025 marketing + support + transaction/risk losses = $635.384m + $457.978m + $407.323m = $1.501bn, or 68.6% of $2.187bn revenue.
The interchange sensitivity table assumes unchanged Q2 2026 purchase volume and applies a hypothetical decline in payments-revenue intensity. It is an illustrative scenario, not a forecast or estimate of a likely regulatory outcome.
Sources
1. Chime Q2 2026 Form 10-Q — Link
2. Chime Q2 2026 earnings release — Link
3. Chime 2025 Form 10-K — Link
4. Revolut Annual Report 2025 — Link
5. Revolut Group 2025 Annual Report PDF — Link
6. Monzo Annual Report 2026 — Link
7. Monzo Annual Report 2025 — Link
8. Nu Holdings Q2 2026 results — Link
9. Nu Q2 2026 managerial P&L — Link
10. Nu Holdings 2025 Form 20-F — Link
11. Federal Reserve debit interchange data — Link
12. Federal Reserve small-issuer exemption — Link
13. Federal Reserve Regulation II — Link
14. UK PSR Interchange Fee Regulation — Link
15. UK PSR cross-border interchange review — Link
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