Open banking arrived in Britain as a competition remedy. Regulators had concluded that the largest retail banks enjoyed an advantage that customers could not easily escape, largely because switching was difficult and because the incumbents held data that nobody else could see. The solution was to force the banks to share, with customer consent, through standardised interfaces. The expectation was a wave of new competitors.
What happened instead was quieter and more interesting. Open banking did not produce a challenger that displaced the major banks. It produced infrastructure — plumbing that now sits invisibly beneath a large number of financial products, most of which the user does not think of as open banking at all.
The original theory and where it missed
Policymakers imagined an app that would show a customer all their accounts in one place, compare them intelligently, and recommend better products. Several such apps were built. Most struggled, because aggregation on its own turns out to be a feature rather than a business. Customers were mildly interested in seeing balances together and largely unwilling to pay for it.
The commercially successful uses were different. Verifying income for a loan application in seconds rather than days. Confirming affordability for a rental agreement. Initiating a payment directly from an account without card rails. Reconciling a small business’s bookkeeping automatically. None of these are consumer-facing products in the way regulators anticipated, and all of them work because reliable data access removed friction from a process that already existed.
Payment initiation is the underrated half
Public discussion focused on data sharing, but the second capability — initiating payments on a customer’s instruction — has arguably had more effect. It gives merchants and platforms a route to take money that bypasses card networks and their interchange fees, with strong authentication built in.
Adoption has been strongest where amounts are large enough that card fees hurt: tax payments, deposits, wholesale settlement, investment funding. The Financial Conduct Authority has tracked the growth of payment initiation alongside data access, and the trajectory of the former is the more commercially significant of the two.
Why Britain led and then stalled
Britain had a genuine first-mover advantage. A mandated standard, an implementation entity with authority over the largest banks, and a regulator willing to enforce timelines produced working interfaces years ahead of most jurisdictions. Other countries copied the model.
The stall came from governance rather than technology. The implementation entity was created to enforce a competition order, not to run permanent national infrastructure, and the question of who should own, fund and govern the system long-term proved genuinely hard. Banks resisted paying indefinitely for infrastructure that benefits competitors. Fintechs resisted a governance structure controlled by banks. The HM Treasury has consulted repeatedly on a permanent settlement, and the delay has real costs: uncertain rules deter investment in products built on top.
Reliability is the unglamorous constraint
A payments and data system is only as useful as its worst participant. If one large bank’s interface is unavailable for several hours, every product depending on it degrades, and users blame the fintech rather than the bank. Availability and response-time reporting became a live regulatory issue for exactly this reason.
This is a general lesson about mandated infrastructure. Compliance produces something that technically exists. Commercial dependence requires something that works at three in the morning on a bank holiday, and the gap between those two standards took years to close.
Open finance and the widening scope
The logical extension is to apply the same principle beyond current accounts: mortgages, pensions, savings, investments, insurance. A customer who can grant access to their full financial position enables advice and comparison of a far higher quality than anything available today.
This is where the greatest consumer benefit probably sits, and also the greatest risk. Pension and investment data is more sensitive, more complex and more easily misinterpreted than a current account statement. Work by the Pensions Regulator on data standards illustrates how far the underlying record-keeping in some sectors is from being ready for machine-readable sharing.
Consent is doing a lot of load-bearing work
The entire framework rests on the idea that customers meaningfully consent to data access. In practice consent is granted through a short flow that most users complete without reading, and re-authentication requirements exist partly to force periodic reconsideration.
Guidance from the Information Commissioner’s Office on lawful bases and transparency applies here, but the deeper problem is behavioural rather than legal. A framework that assumes informed consent while relying on flows optimised for completion rates carries an unresolved tension. Nobody has solved it, in Britain or anywhere else.
What maturity looks like
Open banking succeeded in a way that is easy to undervalue because success looked like infrastructure rather than disruption. Millions of people now use it monthly without knowing the term. The next phase depends on settling governance, extending scope carefully, and holding participants to availability standards that make dependence rational.
None of that will generate headlines about bank-slaying startups. It will, over time, make a large number of ordinary financial processes faster and cheaper, which was always the more realistic prize.
