Avoiding a Leak in the Foundation

I’ve lived in my current home for a little over five years. For much of that time, a house around the corner has sat on the market. 

It’s beautiful, with sweeping views and spaces that invite the sun at all angles. In real estate speak it’s described as an “executive residence”. It’s being sold at an “unbeatable price”. Still, it sits on the market.

The problem with it is pretty obvious, and pretty common among New Zealand houses built in the 1990s.

Monolithic cladding.

In the 90s, the New Zealand housing market fell head over heels for monolithic cladding: a shiny new architectural trend. It was cheap, fast to install, and it promised a sleek, seamless Mediterranean aesthetic everyone suddenly wanted. 

But there was a costly catch. The industry jumped to a modern material without analysing how it would respond to the New Zealand climate - a climate seldom described as “Mediterranean.” Because the cladding was fixed directly to the timber frames without a drainage cavity, the tiniest exterior crack allowed water inside. The moisture became permanently trapped, rotting untreated framing from the inside out.

The material gave the illusion of a fast, modern solution, but it built a multi-billion-dollar disaster into our already fragile housing stock.

Because of it, a house sits in my neighbourhood for a bargain price, if you don’t factor in remediation.

Similarly, directives to adopt new technology look to deliver budget savings, without considering the rush to adopt might lead to remediation costs down the line. 

The AI directive

We'll need people on the front line, but if we can support them with better digital tools and with AI - that's the right thing to do.”
- Nicola Willis, Finance Minister, May 2026

The government has made it clear they have an expectation that the public sector needs to lean heavily into Artificial Intelligence and digital tools to drive efficiency. On paper, it sounds great; who doesn't want a smarter, more productive system?

But, in the real world, this kind of top-down mandate often triggers a very predictable, very costly reflex: the urge to use AI simply for the sake of saying we used it. 

If you build your strategy on unchecked assumptions, you end up applying advancing technology to a fundamentally broken system. Rather than resolving the core friction, you simply automate the chaos. You solve the wrong thing with spectacular efficiency.

Stay agnostic, stay curious

This is why staying tool-agnostic is so critical. The moment an internal team or an external partner hitches their wagon to a specific platform or the latest tech trend, their ability to honestly diagnose what a business actually needs goes completely out the window.

Instead of peddling a pre-packaged answer, the focus should be on pure curiosity. It’s about having the discipline to slow down, harness the information gap, and completely understand the "why" before anyone touches the "how".

To get to the true needs of an organisation, you have to look at the reality of three things:

• People: Who is actually doing the work, and what do they need to achieve?
• Process: Is the workflow logical, or is it fundamentally fractured?
• Technology: What tool - if any - actually supports those first two?

Only when you map those out can you properly define the scope. Sometimes the answer really is a sophisticated machine learning model. Just as often, the answer is a simplified workflow, a policy tweak, or a cleaner spreadsheet. The technology should never dictate the strategy.

The projects that survive are the ones where someone does the diagnostic work before anyone gets near a platform. That means sitting with the people doing the work, not just the people commissioning it. It means being willing to come back and say "the problem you described isn't the problem"- and having that received as value rather than obstruction. 

That kind of rigour is hard to maintain when there's momentum behind a digital push. But, if you really want to create efficiencies and allow technology to support - not replace - human judgement, mahi, and intelligence, you need to ask the questions that get to the real problem. It’s obvious, but that’s the only way to get to the real solution.

Learning from a monolithic error

The multi-billion-dollar tragedy of our leaky buildings didn't happen because monolithic cladding was a bad material. It happened because the industry rushed to install a modern tool without asking a foundational question about how it would handle the rain.

When it comes to the current digital push, the risk is exactly the same.

If the directive doesn’t allow the discipline to slow down and find the real problem first, the most sophisticated AI tool in the world won't save you. It will just automate the chaos - and by the time you spot the rot in the framing, the remediation bills will already be due.

Let’s do what we can to make sure we apply the right solution to the problem the first time.

‍

Jamie is Redvespa's Head of Experience. While his morning walk takes him past that house selling for half its potential value, he's more likely to be found asking questions about the right digital tool to use at work than figuring out the physical tools to fix a leaky home.

‍