Why Every Neobank You Use Is Built on Rented Infrastructure - And Why That Matters

The bank on your phone almost certainly does not own the technology that actually moves your money. The app is theirs. The pipes underneath it belong to someone else entirely.

This isn't a secret. It's a business model - and for the first decade of the neobank revolution, it worked remarkably well. Building on existing financial rails allowed companies like Revolut, Monzo, and N26 to launch fast, iterate quickly, and acquire tens of millions of users without the decade-long process of building core banking technology from scratch.

But as we move through 2026, the cracks in this model are becoming harder to ignore. And for anyone trying to understand where banking technology is heading - or where to put serious capital - understanding the difference between rented and owned infrastructure is no longer optional. It's foundational.

How Neobanks Actually Work - The Layer Most Users Never See

When you open a neobank app and tap "Send €200," a chain of dependencies fires that most users never think about. Understanding that chain is the first step to understanding why infrastructure ownership matters.

A typical first-generation neobank operates across five distinct layers:

1. The app layer - owned

   The mobile front-end: the UX, design, notifications, onboarding. This is what most people think of as "the bank." It is the only layer the neobank fully controls.

2. The core banking platform - rented

   The ledger system that records who owns what. Most neobanks use third-party platforms such as Mambu, Thought Machine, or 10x Banking - licensed SaaS products where the neobank pays a fee to sit on someone else's ledger.

3. The banking license - licensed or partnered

   Operating as a bank requires a regulatory license. Many neobanks operated under a partner bank's license for years rather than holding their own. Revolut only obtained its EU banking license in 2021 - four years after launch.

4. The payment rails - third-party

   When money moves between accounts, it travels over networks owned by Visa, Mastercard, SWIFT, or SEPA. Neobanks access these rails through partner relationships. They do not own the tracks.

5. Card issuing - outsourced

   The card you use is almost universally issued through a third-party card processor - companies like Marqeta or GPS. The neobank brands the card; someone else manufactures the financial logic behind it.

The Hidden Costs of Banking on Borrowed Rails

Rented infrastructure enabled fintech innovation at scale. But it comes with structural costs that are increasingly visible as the industry matures.

Cost floors that can't be engineered away

Every platform fee, every per-transaction charge to a processor, every licensing cost for the core banking SaaS - these are permanent line items. They can be negotiated at scale, but they cannot be eliminated. As transaction volumes grow, so do the bills. The margin expansion story that investors expect from software businesses doesn't materialise in the same way when the underlying infrastructure belongs to someone else.

Dependency risk at the infrastructure layer

In 2021, a widespread outage at a major cloud banking platform caused simultaneous failures across multiple UK neobanks. Customers couldn't access accounts. Cards were declined. The neobanks' engineering teams were powerless - the failure lived in the layer they didn't own. This is not a hypothetical risk. It has already happened, repeatedly.

Regulatory exposure through third parties

As regulators in Europe, the UK, and the US tighten oversight of Banking-as-a-Service providers, neobanks that depend on partner banks for regulatory permissions face upstream exposure. Several US neobanks have had to pause or restructure operations when their BaaS partner came under supervisory pressure.

The differentiation ceiling

If your core banking engine is the same commodity SaaS product used by dozens of competitors, your differentiation lives entirely in the app layer. And the app layer is the most rapidly commoditised part of the stack. Competing purely on UX is a race with no finish line.

Infrastructure Ownership Across the Major Neobanks

Here's how the major neobanks compare on infrastructure ownership as of early 2026:

Infrastructure status based on publicly available information, Q1 2026. The industry is evolving.

The Scale of the Problem

The scale of financial exclusion in emerging markets makes the infrastructure question especially acute. An app-only bank requires a smartphone, a data connection, and an OS capable of running the software. In high-growth markets across Sub-Saharan Africa, Southeast Asia, and parts of Latin America, these prerequisites structurally exclude a significant share of the intended customer base.

This is not a niche problem. It is the defining constraint of first-generation mobile banking - and the central reason a second generation is now emerging.

What a Second-Generation Neobank Actually Looks Like

The term "second-generation neobank" is beginning to appear more frequently in fintech circles - but it's worth being precise, because it gets used loosely.

A genuine second-generation approach doesn't simply mean a neobank that has matured and obtained its own license. That's evolution within the same model. True second-generation thinking means rethinking the infrastructure layer entirely - asking not "how do we build a better app on top of existing rails?" but "what if we built the rails?"

1. Proprietary core banking engine

   The transaction ledger, account management, and financial logic are owned internally - not licensed from a SaaS provider. This eliminates the cost floor, removes dependency risk, and creates a technology moat that takes years to replicate.

2. Own regulatory standing

   A banking license held directly - not borrowed from a BaaS partner. This provides regulatory stability and long-term strategic independence from third-party decisions.

3. Hardware integration

   The most radical departure from the first generation. Rather than relying on consumer smartphones designed primarily for entertainment, a second-generation approach considers whether the device itself should be purpose-built for banking - hardware-level encryption, dedicated security chips, the bank as primary function rather than one app among thousands.

4. Offline-first capability

   If infrastructure is owned end-to-end, it becomes technically possible to design transactions that don't require a live internet connection - something architecturally impossible for banks dependent on cloud-based third-party systems. Money that works when the internet doesn't.

Why Infrastructure Ownership Is the New Moat

For investors evaluating fintech in 2026, the infrastructure question has become a primary diligence point - not a secondary consideration.

The first generation of neobanks created enormous value by demonstrating that consumer banking could be rebuilt from the UX layer. Revolut's valuation trajectory validated that thesis comprehensively. But in a market where UX is largely commoditised, the next wave of value will come from companies that own something harder to replicate.

1. Unit economics that improve non-linearly

   A rented infrastructure bank pays more in absolute terms as it scales. An owned infrastructure bank amortises its fixed technology investment across a growing user base - the marginal cost of the millionth user is structurally lower.

2. Data sovereignty

   When the core banking engine is proprietary, so is the transaction data it generates. In a world where financial data is the primary input to credit decisions and personalised services, data ownership is a compounding asset.

3. Licensing and white-label optionality

   A proprietary banking stack can be licensed to other institutions - a revenue stream unavailable to banks that are themselves licensees. Starling's Engine product is an early example. A fully owned stack creates strategic optionality at scale.

4. Defensibility against Big Tech

   Apple, Google, and Amazon are all expanding in financial services. Their primary vector is at the app and payments layer - exactly where first-generation neobanks built their differentiation. A bank with owned infrastructure at a deeper level has a more defensible position against this threat.

Frequently Asked Questions

What Comes Next

The first generation of neobanks changed what consumers expect from a bank. Faster. Simpler. No branches, no paper. That revolution is complete - and the improvements it delivered are now table stakes, not differentiators.

The second generation will change what a bank is. Not a better app on top of the same pipes, but a fundamentally different architecture - one that owns the pipes, owns the device, and operates independently of third parties whose interests may not always align with the bank's users.

The transition is already underway. The question for investors, regulators, and the customers who depend on these systems is simply: which banks are building the infrastructure of the next decade - and which are still renting it?

That distinction will define the competitive landscape of mobile banking for the next ten years.

From the MobiBank Insights

Published by MobiBank® (Helsinki, Finland). Infrastructure status of third-party companies is based on publicly available information as of Q1 2026 and may change. This does not constitute financial advice or an offer to invest. For investment information, visit mobibank.fi/series-a.