$33M IT Project Successfully Converts Money Into Nothing

Nigel
Nigel is the founder, editor-in-chief, and lead writer at Pavlova Post, a New Zealand satire publication covering national news, local chaos, weather drama, politics, transport mishaps, and everyday Kiwi life – usually with a generous layer of exaggeration.

Immigration New Zealand has completed one of the public sector’s most ambitious digital transformations: spending seven years and approximately $33 million transforming a technology project into absolutely nothing.

The Biometric Capability Update project began in November 2018 with the modest goal of modernising Immigration New Zealand’s identity-management system.

By November 2025, it had achieved something much rarer.

It had been discontinued without delivering any measurable benefits.

That is not easy.

Most failed workplace projects accidentally produce a spreadsheet, a half-working login screen or at least a branded reusable drink bottle.

This one reportedly cycled through at least 12 project managers, three programme or project directors and six project sponsors while maintaining an impressively consistent commitment to not arriving anywhere.

The handover process appears to have worked beautifully

Twelve project managers across seven years means the project received a fresh pair of hands roughly every seven months.

The system may not have been capable of reliably managing biometric identities, but it became highly experienced at identifying the next person expected to explain it.

A typical handover may have gone something like this:

“So, where are we at?”

“We are currently reviewing the revised requirements arising from the previous review of the earlier revised requirements.”

“And when does it go live?”

“That depends.”

“On what?”

“Whether you are still working here by then.”

The review found the project also experienced widespread staff turnover across most professional disciplines, alongside significant turnover at governance level.

In other words, the most successfully automated feature was apparently the revolving door.

More than 170 changes added to project that still produced no change

The project reportedly accumulated more than 170 operational-level change requests.

That is a spectacular amount of changing for something that ultimately changed nothing.

For ordinary New Zealanders, a change request might involve asking the builder to shift a power point or telling a café they forgot the trim milk.

In a major government technology project, it can mean altering the scope, budget, timeline, requirements, solution, delivery plan and emotional definition of the word “soon”.

The project began as an upgrade before expanding into a full replacement of the existing system.

The review found that major shift occurred without adequate analysis, detailed business requirements or proper testing of whether the proposed product could actually do what Immigration New Zealand needed.

This is broadly similar to buying a caravan before deciding how many people must sleep in it, then spending seven years modifying it into a ferry terminal.

Red flags remained visible but were treated as decorative bunting

Delivery problems emerged early.

Deadlines slipped. Requirements remained unclear. The proposed solution needed extensive customisation. Testing never properly stabilised. Fixes reportedly introduced fresh defects, allowing the project to generate problems with the renewable enthusiasm of a hydrangea.

Independent reviews repeatedly raised significant concerns.

One review in 2023 reportedly rated the project as unlikely to succeed and questioned whether it should continue.

Officials later provided the immigration minister with advice claiming the project approach was sound and robust, the build was achievable and risk management was effective.

The independent review found that advice was incorrect and misleading.

It found no evidence of a deliberate intention to mislead, suggesting the information simply completed the normal government journey from ‘deeply concerning’ to ‘tracking well’ while moving between documents.

Immigration Minister Erica Stanford said her trust and confidence in officials had been dented after discovering that the cheerful briefing did not resemble the underlying quality-assurance material.

This has created a serious administrative challenge because ministers traditionally rely on briefings to discover what officials would like ministers to believe is happening.

Even the price appears to require its own investigation

MBIE has publicly described the failed project as costing $33 million.

However, the independent review said establishing an accurate total had proved difficult. It referred to estimates of approximately $35 million for the Biometric Capability Update and another $8.5 million for the related In-Person Enrolment project, while earlier records suggested the main project may once have approached $40 million.

Some staff reportedly described elements of the financial management as ‘creative accounting’, including efforts to remain below a $35 million threshold that would have triggered greater cabinet scrutiny.

There is something almost artistic about a project whose cost is no more settled than its requirements.

The final amount may depend on whether the government counts the related work, transferred funding, unfinished components, consultants, reviews, apologies and industrial quantity of meeting-room coffee required to keep saying ‘delivery remains feasible’.

The project has now entered its most productive stage: investigations

The Public Service Commission will investigate integrity concerns involving the quality of official advice and financial management.

MBIE is also conducting a stocktake of its other major technology projects, presumably beginning with the important question:

‘Are any of these also quietly spending millions of dollars while producing a series of colour-coded updates?’

The ministry says Immigration New Zealand’s existing identity-management system continues to operate normally and New Zealand’s border remains safe.

That is reassuring.

The old system is still working, the failed replacement has been stopped, and taxpayers have received the traditional public-sector technology package: