Lead, Follow, or Merge

01Three positions, no fourth

Three positions are open to a mid-tier customer-owned bank over the next eighteen months: lead, follow or merge. There is no fourth.

The fourth position, carrying on with the current platform indefinitely, has quietly closed. Three changes have closed it.

The merger wave. Mutuals on modern platforms are becoming acquirers. Mutuals on legacy platforms are becoming targets. The sector's most active boardroom conversations are no longer about technology costs. They are about technology as the factor that decides who absorbs whom over the next five years. APRA counted 56 customer-owned banks at the start of 2025 and 52 after four mergers that year, and expects the trend to accelerate in 2026.

Operational resilience regulation. APRA's CPS 230 has been in force since July 2025. It requires boards to keep critical operations within tolerance through severe disruption, and to manage the risk in their material service providers. "Our incumbent will tell us when" is no longer a defensible board posture.

The regulatory load itself. Consumer Data Right obligations are widening, payment rails are evolving, and anti-money laundering reform is arriving. Every regulatory change billed per customer by an incumbent vendor widens the cost gap to institutions on shared, API-first infrastructure: platforms whose architecture exposes every banking function as a programmable interface from day one. The next two years of regulatory change alone are material.

The Standstill, waiting for someone else to go first, was a rational strategy when nothing was changing. The conditions have changed.

Section 07 sets out a ninety-day action ladder for any of the three positions. If you read only one more section, read that one.

02What the Standstill costs

Boards are weighing the wrong risk. The question usually asked is: what if a transition fails? The question that matters is: what does the Standstill cost each year, and who is paying for it?

A failed transition can be recovered from. The Standstill compounds. Every year on a closed platform that resists integration adds:

• A widening cost-to-income disadvantage. The mutual sector's cost-to-income ratio reached 78.15% in FY2025, up 135 basis points in a year, against roughly 45 to 50% for the major banks.
• Slower product release cycles, measured in months where API-first equivalents are measured in weeks, which makes parity on origination, servicing and digital experience impossible to hold.
• A per-customer rebuild of every regulatory change, where comparable platforms absorb the change centrally and pass the result through at no extra cost to each customer.
• Concentration in a small number of incumbent vendors, with no contingency a regulator would accept as adequate under CPS 230.

The deferred decision is the more expensive decision. It does not appear on this year's profit and loss statement, which is exactly why it is dangerous.

03The eighteen-month window

The Australian mutual sector now has a public data point. In August 2026, a first-mover mutual completed a full core banking migration to a modern platform, after an eighteen-month program, and the move has been widely reported as smooth.

That opens a second-mover window of roughly eighteen months. After that, the pattern hardens. A small group of mutuals will have moved, and the rest will face a steeper path, because the available talent, the proven implementation patterns and vendor capacity will concentrate around the early movers.

Eighteen months is enough time to:

• run a structured discovery (eight to twelve weeks)
• evaluate and select a vendor (twelve to sixteen weeks)
• mobilise a delivery program (four to six weeks)
• reach a first production milestone on a single product line, with the new platform running alongside the legacy one (nine to twelve months from mobilisation)

It is not enough time if the board decides to begin late in the window. To use the second-mover window, the board decision needs to be made in the next six months.

“

Three positions are open. There is no fourth. The Standstill is now a path to position three.

04Three bars for a viable alternative

If the sector is to move, the alternative has to clear all three of these bars.

1. Materially better unit economics. A 5 to 10% saving will not move a board, because the implementation risk premium will absorb anything that small. Sector practitioners put the threshold for a viable alternative at around 20 to 30% below the current run rate, demonstrable in the first year of steady state, and well before year five. Mutual cost-to-income ratios have no room for vendor margin shaped the way the major banks tolerate.

2. Open by design from day one. The next platform has to be API-first as a design principle. Modern payment rails, customer data infrastructure, identity services and partner integrations need to be available on day one, rather than promised on a roadmap. The mutual sector cannot afford another walled garden.

3. Regulatory change absorbed centrally. When the next regulatory change of the scale of Consumer Data Right arrives, it has to be rolled out once, across the platform's whole mutual customer base, rather than billed to each customer. Over five years, this is the largest single difference in total cost of ownership that the sector has not yet priced.

A cultural test sits behind all three: the vendor's economic model has to tie its success to the success of its mutual customers. If a vendor profits when its customer struggles, the relationship will not survive a five-year program.

05Why core banking migrations fail

Boards in the mutual sector carry memories of core banking programs that went badly. The most thoroughly documented failure in modern banking is the 2018 migration of TSB Bank in the United Kingdom to a new core platform. The independent review by law firm Slaughter and May, a 262-page report, identified three causes, each of which can be addressed:

A big-bang cutover, without proper consideration of alternatives. The migration was structured as a single weekend event, and other approaches were not given real consideration. When it failed, a significant proportion of TSB's 5.2 million customers were affected, and the bank did not return to business as usual until December 2018. UK regulators fined TSB £48.65 million, and it paid £32.7 million in customer redress.

No proper independent assessment of the vendor. The migration was run by the IT supplier owned by TSB's parent company. The review found that TSB did not properly assess the supplier's capability before the project started, or manage it at arm's length once it had. The supplier's track record was taken on trust.

No independent technical assurance. The review found there were no expert external advisers for the program as a whole. The board had no independent view challenging the supplier's readiness assurances.

These failure modes can all be designed out. They are arguments about how a migration is structured. Avoiding each one is a matter of program design, whatever the technology, the vendor or the sector.

06The coexistence-first migration model

Our team has led core platform transformations at major Australian banks, including core banking modernisation, a $9.6B account migration, a $2B+ banking simplification program and a $430M wealth management separation. The mutual context differs in scale, capital position and member relationship, and the method that consistently takes the risk out of a core platform transformation carries across. It has four moves.

Build new alongside legacy from day one. The new platform stands up in parallel. New products and new customers go on it first, while the legacy platform keeps serving the existing book without disruption. If the new platform underperforms under real load, the existing book is unaffected.

Prove it at small scale before any forced migration. A new product line, a single customer segment or a new acquisition channel becomes the proof point. Broader migration is scoped only once the platform has worked in production, with real customers and real volume.

Migrate in tranches. Customer groups move in waves over weeks, rather than in a single weekend cutover. Each wave is a learning event, and issues found in one inform the next. The institution keeps the option to slow down or pause at every tranche boundary.

Decommission last. The legacy platform retires only after the new one has been proven across every product, segment and edge case. Months of overlap cost real money, and they are the cheapest insurance the program will buy.

The discipline underneath all four moves is rehearsal. When Australia's largest bank moved its core banking platform to the cloud, the program ran for eighteen months, the team rehearsed the cutover five times against three planned, and the bank was fully offline for only three hours on the day. Few mutuals can fund a program at major-bank scale, and they don't need to. The method is a sequencing and rehearsal discipline, and it works at any budget.

What the Merge position actually looks like

The Merge position can be the strongest move available. For some mutuals, particularly those whose member base, geography or scale make standalone modernisation uneconomic, a planned merger into a larger, modernised institution is the best outcome for members. The question that decides it is whether the merger happens by intention or by force.

• By intention, you choose your partner and negotiate the terms. Your members receive the benefits of a modernised platform without the program risk, and your leadership team negotiates from strength.
• By force, you defer the platform decision until it becomes unaffordable. Members get worse outcomes than under either standalone modernisation or a planned merger, and leadership negotiates from need.

Institutions in the Merge position need that conversation now.

07The ninety-day action ladder

A board ready to act on any of the three positions has three concrete steps available in the next ninety days. None commits the institution to a vendor, a platform or a budget. All of them produce a decision the board can stand behind, and they sharply reduce the political exposure of the CEO.

Step 1: Commission an independent assessment. Four to six weeks, fixed scope, fixed price. It covers the institution's current platform position, its modernisation options, and the risk-adjusted economics of acting against deferring. Output: a board paper with a recommended path, decision points and a quantified risk envelope.

Step 2: Develop a tranched migration architecture. Six to eight weeks, specific to the institution's product mix, customer base, capital position and operational limits. Output: a coexistence-first program design, a phased capital plan and a critical-path schedule.

Step 3: Approve a coexistence-first proof of concept. Three months to first production on a single new product line or segment, with no impact on the existing book. Output: a working production instance of the new platform alongside the existing one, with measured cost, performance and customer outcomes.

Three months. A clear board decision. A first move that cannot fail catastrophically.

The independent adviser's job across these steps is specific: pressure-test the business case before any commitment is made, bring the lessons of successful and failed core programs into the mutual context, sequence the program to the institution's risk appetite and capital, hold the vendor to the three bars throughout, and reduce the political exposure of the decision so the CEO does not carry a single point of failure. This work takes months, and it happens before vendor selection.

08A closing word

The Standstill will end. The only question is how.

The orderly path is one or two institutions proving a viable pathway and others following. The disorderly path is the cost gap and technology debt forcing a less considered response across the sector. The first is achievable inside the next eighteen months. The second is the path to position three.

What would need to be true for your board to begin this conversation in the next six months?

Would it be unreasonable to compare your current five-year trajectory against the institutions that have already moved?