Universal banking was built on the assumption that breadth is strength: more products, more markets and more capabilities under one brand. Decades of acquisitions and in-house builds have left many large banks running as loose federations of partly integrated platforms, and regulators on both sides of the Atlantic now ask a different question of them, which is not how much they earn but whether they can stay in control under stress.
The contributed piece below argues that complexity has stopped being a cost line and become a resilience problem, that AI adoption will make it worse where the foundations are weak, and that the banks which win the next decade will be simpler rather than bigger.
Tej Patel is a partner at Elixirr, a consultancy that works with banks and other financial institutions on strategy, technology and operating model change. The article that follows sets out his opinion.
For bank executives, the pressure to move faster has rarely been higher.
Boards want growth. Investors want efficiency. Customers want simpler, faster and more intuitive digital experiences. Regulators want stronger controls, clearer accountability and greater operational resilience. Across the industry, there is also a very visible expectation that banks should be moving quickly to adopt AI and unlock productivity gains.
None of these demands is unreasonable. The challenge is that they are all landing at the same time, on organisations that are already carrying decades of accumulated complexity.
For many banks, the issue is not a lack of strategic intent. Most have a clear view of where they need to go: digitise the customer experience, improve productivity, modernise technology, strengthen controls, use AI more effectively and reduce cost. The harder question is whether their operating model can absorb and execute that agenda without becoming more fragile in the process.
The uncomfortable truth is that many banks are trying to move faster without first becoming simpler.
That matters because complexity is no longer just a cost problem. It is a resilience problem.
Complexity is becoming a strategic constraint
For years, scale and breadth were seen as sources of advantage in banking. Broader customer relationships, diversified revenues, global platforms and the ability to serve more client needs under one roof all had clear strategic logic.
That logic has not disappeared. But unmanaged universality is becoming harder to govern.
Decades of acquisitions, product expansion, regional growth and attempts to own every capability internally have left many banks operating as federations of partially integrated platforms behind a single brand. In stable conditions, that complexity is expensive and inefficient. Under stress, it becomes a control and resilience issue.
This is where the trade-off has shifted.
The question is no longer simply whether large, complex banking models generate adequate returns. It is whether they can be governed, changed and protected at the speed now required.
For many banks, the next source of competitive advantage will not be more scale. It will be the ability to simplify without losing relevance.
That requires patience and discipline, which can be difficult in an environment where executive teams are under constant pressure to deliver near-term outcomes. Simplification is not always immediately visible to the market. It often means retiring platforms, rationalising processes, reducing duplication, clarifying ownership and making hard choices about where the bank genuinely wants to compete.
But without that discipline, transformation activity can easily add another layer of complexity rather than remove one.
AI is increasing the urgency and the risk
AI is now one of the clearest examples of this tension.
Many banking leaders are asking the right question: how do we accelerate AI adoption and use it to drive efficiency, improve decision making and enhance the customer experience? The pressure to move is real. The market expects it, boards are asking about it, and competitors are experimenting at pace.
There are undoubtedly high-value use cases across fraud detection, risk monitoring, customer servicing, software development, operations and knowledge management. Used well, AI can help banks improve productivity and responsiveness in material ways.
But AI will not fix complexity on its own. In some cases, it risks accelerating it.
If AI is layered onto fragmented data, unclear process ownership, weak governance or legacy architecture, it can create additional operational and control challenges. Banks may automate activity without simplifying the underlying process. They may generate productivity in one part of the organisation while increasing validation, monitoring and governance burdens elsewhere. They may move quickly on pilots but struggle to scale because the foundations are not ready.
The issue is not whether banks should adopt AI. They should. The issue is whether AI adoption is being treated as part of a broader resilience, technology and operating model agenda, or as another standalone transformation priority competing for the same scarce technology and change capacity.
Too often, transformation priorities are already competing for the same teams. Digital ambitions are being slowed by legacy platforms and fragmented data. Resilience is considered too late in the programme lifecycle, once key decisions about architecture, suppliers, data, controls and operating processes have already been made.
That sequencing has to change.
Resilience cannot sit behind the transformation agenda. It has to shape the transformation agenda.
Regulation is reinforcing the point
Regulators are increasingly focused on whether banks can remain in control under stress. That means understanding which services are critical, who owns them from end to end, where key dependencies sit, how failures are contained, and how quickly the organisation can recover.
This is not simply a compliance exercise. It goes to the heart of how a bank is designed and run.
After crises, rules tend to accumulate. They rarely disappear. Banks often respond by adding new frameworks, new committees, new controls and new reporting layers. Each response may make sense in isolation, but the cumulative effect can be more complexity, not more control.
The stronger response is to simplify governance, clarify accountabilities and embed resilience expectations into the way strategic decisions are made.
That means asking different questions at the start of a programme, not at the end:
- Does this change simplify the organisation or add another dependency?
- Which critical services are affected?
- What new supplier, data or technology risks are being introduced?
- Who owns the outcome from end to end under normal conditions and under stress?
- Will this improve our ability to respond and recover, or make it harder?
These are not just risk questions. They are CEO, COO, CIO, CTO, CRO and board questions.
The leadership challenge is prioritisation
The challenge for many executive teams is not ambition. It is prioritisation.
Most banks have more strategic priorities than their organisations can realistically deliver at pace. Cost transformation, AI adoption, technology modernisation, customer experience, regulatory remediation and resilience are all important. But they draw on the same limited leadership attention, funding, technology capacity and change capability.
When everything is a priority, execution slows. When programmes are launched without retiring old activity, complexity grows. When cost reduction is pursued without understanding operational dependencies, resilience can weaken. When AI adoption is accelerated without fixing data and process foundations, scale becomes difficult.
This is why simplification needs to be treated as a strategic capability, not a tidy-up exercise.
The banks that make progress will be the ones that are clear about where they create value, where they need to partner, what they should stop doing, and which capabilities are genuinely differentiating. They will use AI and automation with discipline. They will modernise technology around clear business outcomes. They will involve risk, operations and technology leaders at the point of strategic choice, not just when controls need to be signed off.
That is not easy work. It requires executives to balance immediate performance pressure with the longer-term discipline needed to reduce fragility. But that balance is now central to competitiveness.
The winning banks will be simpler, not just bigger
Operational incidents will continue to happen. Cyber disruption, supplier failure, AI-driven fraud, cloud outages and internal control breakdowns are all part of the risk landscape for a more digital and interconnected banking sector.
The differentiator will not be whether a bank can prevent every disruption. It will be whether it can understand the impact quickly, mobilise across silos, communicate clearly, recover effectively and continue to serve customers without losing control.
That kind of resilience cannot be bolted on after the fact. It has to be designed into the way banks grow, transform and adopt new technology.
The winning model in banking will not be universal scale at any cost.
It will be focused growth around clear strengths, disciplined technology choices, smarter collaboration and a relentless focus on simplifying the organisation while improving customer relevance.
Banks that accept resilience as their new hard limit, and design within it, will be better placed to satisfy regulators, withstand technology-driven shocks and keep pace with customers whose expectations are already moving faster than many banking operating models can support.
The post Resilience, Not Reach, is Banking’s New Hard Limit appeared first on The Fintech Times.
