As part of the Edge 50 list, Apolitical is publishing a series of articles exploring a selection of these bold initiatives to learn more about the stories and strategies behind them.

Created by Apolitical in partnership with the Mohammed Bin Rashid Centre for Government Innovation, this list is an invitation to look beyond the usual playbook and ask what becomes possible when governments are willing to think boldly about today's biggest problems.

Apolitical spoke to Angus Bargh, Founder and Director of Open Plan, an independent company that built and runs the Underground Asset Register in Wellington on behalf of a charitable foundation and the city council. The Register is a shared digital map of the underground pipes, cables and networks that run beneath the city.

In Wellington, missing data meant nearly half of contractors hit unexpected pipes, costing New Zealand's capital city around $50 million a year. The Underground Asset Register changes that, but the real challenge wasn't technical. It was persuading competing utilities, a mix of publicly and privately owned operators, to share data voluntarily in a single place. The result is faster, safer infrastructure delivery, built on a counter-intuitive insight: hidden infrastructure works best when it's visible to everyone. The Wellington business case estimated a 12-to-1 return on investment.

How would you define the core challenge you were trying to solve with the Underground Asset Register?

Wellington had a capital infrastructure investment plan, spanning about 15-20 years ahead. It laid out plans to deliver an unprecedented number of civil infrastructure projects across a range of large-scale initiatives, including new tunnels, water infrastructure and some light rail infrastructure. It was an investment of somewhere between 10 and 15 billion dollars.

Now, Wellington, the capital city of New Zealand, has a population of around 300,000 and is quite a dense, geographically constrained city. This means that to carry out all these plans, the streets that carry our infrastructure, both the transport infrastructure along the top and the lifeline infrastructure underneath (like pipes and cables), were going to need to be dug up many, many times over the coming decade. And that made us think: if we build a database of what we know is underground, then we capture more information as we expose those assets throughout these projects, we'll build this rich library of information that will help to expedite projects in the future, and obviously bring efficiencies and save money decades ahead. So there was a short-term benefit, and also a long-term vision to improve things over time.

How does the Underground Asset Register work in practice? What does it change?

Previously, if a contractor working in the field wanted to understand what would be underground when they excavated, they would have to contact each individual utility and request plans. Sometimes, those plans could take up to a week to be returned, and they would be delivered via email as PDFs. They'd potentially have 10 or 12 different sets of plans, and there was no way of overlaying them. They weren't digital, and they'd have to use their expertise to try to understand what those plans would look like if you overlaid them all. Then, they'd have to validate that understanding. They'd do things like ground penetration radar surveys, and there's a mechanism called potholing, where you dig small holes to actually look and see if the service is where you think it is.

The Underground Asset Register accelerates this process. It means you can bring in data from those utilities ahead of time, and it means we can standardise the way these plans look on a map. Thanks to the Register, the delivery of those plans now takes seconds rather than days. Contractors can also get a better sense of the accuracy of the data being provided to them. This allows them to determine how much of this validation work they actually need to do. Often, they needed to do less. Those savings really compound over time. Our research estimates that poor underground asset records contribute over $1 billion in avoidable costs nationally each year, and that discovering unidentified services during excavation can add up to 20% to project costs. The Wellington business case estimated a 12-to-1 return on initial investment.

What design choices made the project a success?

We adopted a set of principles from the outset. They're called the Gemini Principles. We borrowed them from an initiative in the UK that emerged from a centre of excellence called Digital Built Britain. They're really about building technology that's open by default and open by choice, ensuring what you build is secure and that you involve key parties at every step along the way. It's about centring collaboration in design. Some people think that collaboration can slow you down. We had to figure out ways to move at speed, while still being very collaborative.

What did it take to get agreement on the idea of a single source of truth?

Collaboration was a key part of the project. It required several parties to come together. The project was really a challenge, not so much in technology, but more in actually getting parties to work together in a way they hadn't before. We have established systems for understanding, from an engineering context, where things are, and for getting plans from utilities showing where their networks are and what's underground. But of course, bringing an online map and federating or consolidating all that data together in one place requires those utilities to behave differently. I think the long-term vision, combined with the challenge of getting these projects done here and now, plus ideas of where we can all get to in the future if we do this right, really brought everybody together.

What kind of resistance did you face from utilities around sharing this kind of information?

There are three components here: data quality, data security and commercial sensitivity. Reluctance to share utility data was often due to concerns about data quality. When that comes up, we often explain that presenting data and then disclaiming its quality is actually a much better outcome than not providing any data at all. Contractors would rather see data and then be told, "It's not that accurate, please check with us for further information," especially when it's about underground assets that could pose a risk to life if struck. Security was another point of resistance. We wanted to ensure we didn't grant system access to malicious actors and that the data wasn't used for nefarious purposes. To address this, we implemented a four-tier security system for different assets, from level 0, where anyone can see exactly where they are, to higher levels where we reveal general, not exact, asset locations, or we simply notify the asset owner that a contractor is looking up their area. The last point of resistance was around commercial sensitivity. Some utilities are in competition with each other, so there were concerns about giving a commercial advantage to other vendors. Our security tiers help us address this, allowing people to obfuscate the exact location of assets.

What tactics did you use to consolidate the security levels?

Throughout the programme, we convened a reference group: the Technical Reference Group. So it was a cross-sector spectrum of players, including design engineers, surveyors, government representatives, contractors, and road controlling authorities. Going back to our design principles, we'd share how we were thinking a particular aspect of the Register would work, test with the Group and implement their feedback.

"With any sort of innovation, there has to be a level of iteration... If something doesn't work in practice, we go back, we change, we try again"

That's what we did with the security levels.

What were the main challenges you encountered during this project?

We're bringing change to well-established, well-understood, sector-embedded government processes for understanding where things are and working around them. When you're bringing digital transformation to something like that, the whole thing needs to change end to end, and that's difficult because you need multiple people to change simultaneously. To overcome this, it's really important to talk about your vision and what you'll all be able to achieve if you go on this journey together.

Working in the public environment on civil infrastructure presents lots of challenges. So many actors need convincing. We get lots of noise from commuters if their journeys are disrupted, for example. Often, with change, people are very keen to understand what the impact will be on them. We just needed to give them the confidence that they would still be able to do everything they used to, and a whole lot more.

"Even once you've got a critical mass on board, you can't stop. You have to constantly educate the market. Give them confidence that the change is worthwhile, that it will bring about transformation in their own business and their own way of doing things."

Another challenge was legal frameworks. In the old regime, the entity responsible for the accuracy of those plans would be the utility provider. Under the new regime, this level of responsibility is being taken up by the platform provider (us) and the councils that adopt the platform. So we had to develop a whole legal framework.

Has anything surprised you about the impact of this project?

We built a business case around the people we thought would benefit and the types of activities that would benefit from this kind of technology. What we found in the early stages of the project was that many actors we didn't think would benefit were saying they were. For example, in local government, there are people whose role is to identify assets without clear owners. They started reporting back that the Register was saving them days of time, literally.

Another example is in temporary sports events or temporary concerts. They require stakes to be driven into the ground to support, for example, large marquees. It's incredible how many times driving those metal stakes into the ground strikes an asset. So events teams were now able to quickly identify whether they should take extra precautions due to assets at their location. My advice is to listen carefully to what's going on in the market, because change comes from surprising quarters, not where you expect it.

What other hidden layers of our cities do you think governments should map or design around? And what would it take to make that kind of thinking standard practice?

This is a very exciting space for us. I was involved in the recovery effort after a major earthquake in my hometown of Christchurch. I took many of my learnings from there into this project. I would suggest that when you're delivering civic and civil infrastructure in public spaces, there are three things that you care about. One is the Underground Asset Register. When I break ground, what am I going to find? What's going to be a risk to me and a risk to my project happening on time and on budget?

You also care about who else will be working in the environment at the same time. What coordination effort do I need to invest to make sure this project isn't too disruptive? Who's going to be working upstream or downstream? Am I planning to work on infrastructure where someone else will come back and effectively change what I've done? The third is everything that's non-asset and non-coordination-based, but affects expediting your project. Things like what are the ground conditions when I dig? Maybe there are significant issues for the Indigenous population related to archaeology and traditional sites of interest. Are there notable trees to be protected? Are there public transport routes? Our aspirations go beyond the Underground Asset Register to create a federated environment where all that public data is available in one space, together, for the public good.

For public officials elsewhere considering creating a similar system to the Underground Asset Register, what practical advice would you offer?

I think there are two ways to approach this. One would be to consider what statutory powers or requiring powers you already have at your disposal to make something like an Underground Asset Register work. If you have the required and statutory powers to say, actually, we can compel people to share data, I'd tell you to put that in your back pocket for a period of time, and then take the second route, which is the collaborative route. So you invite people to contribute voluntarily, because, in our experience, certainly in a New Zealand context, voluntary, non-mandated, collaborative involvement moves things much faster than a resistant, mandated involvement. One thing we found successful was that parties were quite willing to get on board, share data, and participate because of the collaborative nature we set up for the whole framework.

I would also say that this isn't a waterfall delivery. It's not all or nothing. You have to accept that it's a journey you'll need to go on over a number of years. Define what your priorities are, rank your priorities in order, and just really focus on the one at the top of the list for this quarter. Ask yourself: what can you do today to move this forward? Is there a system we already have that could maybe share data? So, for example, when we go into a new city in New Zealand, we don't say we need all the utility data; we ask, "What's available and what will be most valuable for you to observe?" We focus on those things first. That's the value case. And we know we'll mop up the other utilities over the coming months.