In branch programmes, the most expensive decisions are often made before anyone sees a tender price. I have seen how small assumptions in planning, specification and delivery can compound into rework, delays and long-term maintenance cost. This article looks at why retail bank branch ROI needs to be judged across the full lifecycle, not just CapEx.
Hew Kenn Chew, Head of Delivery, Context Architects
Outline
- Why headline CapEx never matches the final figure
- The four invisible cost lines in branch programmes
- Rework and the cost of getting it wrong
- Programme slippage and lost trading weeks
- Variation claims from fragmented design responsibility
- Long-tail FM costs on bespoke fit-out components
- The 4 to 6 per cent floor area problem
- Building a lifecycle business case your CFO trusts
Key Takeaways
Initial construction cost typically represents only a quarter to a third of a building's total lifecycle cost, a ratio consistently borne out by whole-of-life cost research. Branch business cases anchored on CapEx per site systematically understate the real number, because the largest costs sit in post-fitout alterations, maintenance call-outs, rework, fragmented purchase orders, and floor area inefficiency. Construction Industry Institute research puts rework alone at an average of 5 per cent of project value, reaching as high as 20 per cent on poorly integrated programmes, and the gaps between disciplines in a fragmented delivery model are precisely where that cost originates. A defensible business case starts with lifecycle thinking, not square-metre rates.
- CapEx is roughly a quarter of true lifecycle cost
- Rework adds 5 to 20 per cent to project budgets
- Every closed week erodes branch trading revenue
- Post-fitout alterations thrive in fragmented delivery models
- Bespoke fit-out components inflate long-term FM costs
- Spatial inefficiency leaks 4 to 6 per cent rent
- CFOs trust whole-of-life numbers, not per-metre rates
- Ask one question: what does ownership cost over ten years
Introduction
Most bank branch business cases in New Zealand are written backwards. They begin with a target CapEx per square metre, add a contingency, attach a delivery programme, and present a number for executive approval. Twelve months on, the same business case often looks unrecognisable. We often see the corporate fit-out is two million dollars over, the branch is six months late and the Facilities Management team is quietly carrying a maintenance burden nobody modelled at approval.
This is rarely a competence problem. It is often a systemic problem developed over time. Decades of facilities management research, codified in resources such as the Whole Building Design Guide life-cycle cost methodology, point to a consistent conclusion: initial construction cost represents only a quarter to a third of a building's total lifecycle cost, with some specialist research placing the figure as low as 10 to 20 per cent. The remaining majority surfaces in operations, maintenance, parts sourcing, FM call-outs, rework, programme slippage and floor area inefficiency.
If you are pressure-testing a branch refurbishment business case right now, the right question is not what does it cost to build. It is what does it cost to own across a 7 to 10-year lease. This article walks through the four invisible cost categories that decide the real answer.
The CapEx illusion
The number that lands in the executive summary is almost never the number the bank actually pays. CapEx per site is a useful planning anchor, but it captures only the visible portion of programme cost. It does not capture the variation claims that arrive at month nine, the FM call-outs that accumulate from year three onwards, the cost of reopening a closed branch a quarter later than promised, or the slow accrual of leased floor area that turned out to be unused.
The cost picture across a typical 7 to 10-year branch lifecycle is closer to an iceberg than a single line item. The submerged portion is where the real money sits, and it rarely appears in the same papers that approved the project. Property, Finance, Procurement and Facilities Management each hold a fragment, and the executive committee almost never sees them assembled into one number.
The four invisible cost lines
There are four cost categories that quietly determine whether a branch programme delivers on its business case, and that are almost always missing or under-modelled in the approval papers.
1. Rework
Independent industry research puts the cost of construction rework at between 5 and 20 per cent of total project value, and some labour-hour studies suggest rework can account for as much as a third of on-site activity. In a fragmented delivery model, the interfaces between architects, project managers, quantity surveyors, sub-consultants and main contractors are precisely where rework originates. Drawings produced in isolation, decisions made without cost engineering at the design stage, and unresolved technical interfaces all surface as rework once trades arrive on site.
2. Programme slippage
Every week a branch is closed beyond its planned reopening represents lost trading revenue and a delayed return on capital already deployed. A six-week supply chain delay on a long-lead item, multiplied across a national rollout, can defer revenue capture by months. Slippage rarely surfaces as a discrete line in the original CapEx model, yet it often consumes more value than the savings achieved during cost engineering.
3. Variation claims
In a multi-consultant model, the gaps between disciplines are where variations live. When an architect's drawing does not align with the structural engineer's brief, or when a security requirement was not surfaced to the electrical engineer, the resolution arrives as a variation claim. Variation volume is one of the cleanest indicators of how integrated, or how fragmented, a delivery model genuinely is.
4. Long-tail FM cost on bespoke fit-out
Custom joinery profiles. One-off light fittings. Bespoke counter details that match nothing else in the network. Every non-standard component becomes a future Facilities Management cost. When the original supplier exits the market, or the Auckland branch needs a replacement that does not exist in the Wellington kit, FM sources replacements at a premium and the bank pays year after year.
Why the 25 to 35 per cent rule changes the maths
The lifecycle research is unambiguous. Whether the figure cited is the 25 to 35 per cent range that anchors most life-cycle cost analysis work, or the 10 to 20 per cent figure used by building lifecycle specialists, the implication is identical. Construction is the smaller portion of the bill. Operations, maintenance, replacement and renewal carry the larger portion, accruing across the 7 to 10-year fit-out lease.
For a bank operating a 150-branch network on a rolling refresh cycle, even a modest reduction in long-tail FM cost compounds rapidly. The decisions that drive that cost are made in the initial design phase, not in year three or year seven.
The 4 to 6 per cent floor area problem
Commercial fit-out rates in New Zealand currently sit in a broad band, with retail fit-out work typically priced between $2,500 and $4,500 per square metre and high-specification banking environments often well above that range. Layered on top of CapEx is the recurring rental cost on every square metre the bank leases.
Most legacy bank branches still carry 4 to 6 per cent of floor area that no longer serves a purpose. Unused vault antechambers from a transactional era. Oversized back-of-house circulation. Storerooms sized for paper banking. Pulling that floor area out of the brief at feasibility stage reduces both CapEx and the rent payable across the entire lease. It is one of the highest-leverage cost decisions in a programme, and invisible to a business case anchored purely on cost per square metre.
A practical framework for a lifecycle business case
A defensible business case is not a longer spreadsheet. It is a shared model that brings CFO, Property, Procurement and Facilities Management into the same numbers. Four steps move a business case from CapEx anchored to lifecycle anchored.
First, model the full 7 to 10-year cost envelope, not the build alone. Include projected FM cost based on the proportion of bespoke versus standardised fit-out, expected rework, and rental cost across the lease. The Building Research Association of New Zealand and the New Zealand Institute of Quantity Surveyors both publish frameworks that support whole-of-life cost modelling.
Second, model floor area as a variable, not a constant. Engage the design team early on spatial efficiency, and bring leasing into the conversation before the head lease is signed.
Third, build a variation contingency that reflects the delivery model. Fragmented multi-consultant programmes need a larger contingency than integrated single-provider ones. Pretending otherwise produces year-one budget breaches that erode credibility for the rest of the programme.
Fourth, build the FM lifecycle assumption into the specification. Standardised fittings and a written, network-wide property specification reduce long-tail FM cost more reliably than any contractual mechanism applied later.
The single question that exposes most branch business cases
If a business case answers one question well, it should be this: across a 7 to 10-year lease, what is the all-in cost of owning this branch, not the all-in cost of building it. Most business cases approved by New Zealand bank boards today do not answer that question. They answer the easier one, and they are systematically optimistic as a result. Heads of Property who arrive at the executive committee with a lifecycle model, rather than a square-metre model, are far better positioned to defend their numbers when the variations and the FM costs start to land.
Next Steps
If you are working through the financial case for a branch refurbishment or a wider network programme, the next post in this series unpacks how leading banks evaluate and shortlist branch design and delivery partners using a procurement framework built around these lifecycle realities. For further reading on long-term branch deployment in the New Zealand banking sector, the Context commercial portfolio offers practical case examples. To benchmark your current cost model against the wider New Zealand market, the Context team is available for a confidential discussion via https://context.nz/contact/.
Jump to the other blogs in this series here:
Why the New Zealand Bank Branch Is Being Repositioned (Not Retired) in 2026
Blog Author Bio - Hew Kenn Chew
Hew Kenn Chew brings a practical, lifecycle-focused perspective to retail bank branch design, property planning and programme delivery. As Studio Principal, his work sits at the intersection of commercial architecture, retail environments, masterplanning, compliance pathways and long-term building performance. Hew writes for property, procurement, facilities management and finance leaders who need branch networks to perform beyond opening day. His perspective is shaped by the realities that often decide return on investment - whole-of-life cost, design coordination, rework, standardisation, operational maintenance and the handover between build teams and facilities teams. In these articles, Hew explores why successful branch programmes are not simply about creating better spaces, but about making better decisions earlier so capital investment continues to deliver value across the full lease lifecycle.
Hew Kenn Chew, Head of Delivery & Group Studio Principal Context Architects