The first generation of UK neobanks launched between 2015 and 2017 with a broadly shared thesis: legacy bank customer experience was poor, branch-based distribution was an unnecessary overhead, and a well-designed mobile application combined with a prepaid card product could acquire retail customers at a fraction of the cost of a traditional current account relationship. On the customer experience thesis, the neobanks were largely right — the comparison between an app-native spending notification arriving in real time and a month-end paper statement was stark enough that customer acquisition was fast and cheap in the early years. On the economics of the full banking relationship, the decade has been more instructive about what is hard.
The structural challenge that neobanks have been working through since roughly 2019 is the profitability of retail deposits. A current account relationship that generates interchange revenue from debit card transactions, and perhaps a small net interest margin from overnight float, is not a profitable relationship at scale without either premium subscription revenue or credit products that generate net interest income. The neobanks that have reached genuine profitability have done so primarily through lending — personal loans, buy-now-pay-later facilities, or in the case of the SME-focused neobanks, business lending products. The customer experience innovation was real, but the underlying financial model still required credit economics to work.
What the neobank decade has clarified, from our perspective as investors, is the distinction between customer relationship ownership and infrastructure capability. The neobanks are excellent at customer relationship ownership — mobile UX, app-native product design, low-friction onboarding — and have invested accordingly. What they are systematically less good at is the plumbing that sits beneath the customer relationship: the real-time transaction monitoring infrastructure that scales without false positive degradation, the multi-currency treasury management layer for their SME customers, the credit decisioning models that can profitably underwrite thin-file borrowers, the compliance automation that makes regulatory reporting tractable at multi-million customer scale. These are genuine infrastructure problems, and the neobanks that have invested most seriously in solving them have done so by building internal infrastructure teams or acquiring infrastructure companies rather than by deploying the mobile-first consumer product philosophy that drove their initial growth.
The infrastructure investment opportunity that the neobank decade has revealed is precisely in the layers that neobanks found hardest to build. Transaction monitoring at scale — where a financial institution processing millions of transactions daily needs to catch genuine suspicious activity without generating alert volumes that overwhelm a human review function — is a machine learning problem that requires both domain expertise and significant training data. Fennech Financial, which we backed in 2023, is building real-time transaction monitoring infrastructure for exactly this use case: financial institutions that need monitoring capability above what a basic rules engine provides but are not large enough to justify building a bespoke ML platform. The neobank scaling experience has created demand for this infrastructure that did not exist in the same form five years ago.
The second-order lesson is about the importance of the regulatory relationship. Several neobanks that scaled aggressively on customer acquisition found themselves with supervisory expectations — around financial crime controls, customer due diligence, and operational resilience — that they were not fully prepared to meet at the scale they had reached. The resulting remediation programmes were expensive, disruptive, and in some cases led to restrictions on new customer onboarding at precisely the moment when the business model required continued growth. The neobanks that avoided this pattern invested in regulatory infrastructure proportionate to their customer growth rather than lagging it. That is the lesson we apply when evaluating early-stage fintech infrastructure companies: regulatory investment should scale with commercial growth, not follow it.