What if you paid for water like you pay for power?
Picture your power bill folded into your annual rates, a fixed number with no real connection to how much you actually used. Run the heat pump at 23 degrees all day or barely touch it, the bill stays the same either way. Under that system, there is little reason to bother turning things off or to be conscious of what is being used.
That thought experiment is close to how a large portion of New Zealand handles water. Of our 67 territorial authorities, 45 fund water entirely through rates, so it becomes a fixed annual cost with no link to how much comes out of the tap. A handful run a hybrid model, charging extra once usage passes a threshold. But for most people, nothing tells them how much water they are using week to week, and that is largely why we have never had reason to think about it.
The case for using less
That gap in information matters more than it might seem. Reducing consumption is one of the biggest levers we have for bringing down the cost of three waters infrastructure. Lower usage defers capital upgrades and eases the strain on existing assets, so a $20 million treatment plant expansion can be pushed back simply by freeing up capacity through demand.
Councils that have introduced water charging have seen demand fall by 20 to 37 percent on average. The interesting part is that the drop comes from behaviour change, and what happens once people can see what they're using and start adjusting their habits. Tauranga is a good example. When metering and billing began there, peak daily demand fell by 30 percent and average demand by 25 percent. It took seventeen years for total network demand to climb back to pre-metering levels, which meant the city could add 40,000 people to the scheme without adding capacity. So, when we hear that a network needs more infrastructure to cope with demand, smarter water use is often part of the answer.
The flow-on effects go beyond the supply network too. Less water coming out of our taps generally means less going into the wastewater system. While the relationship isn't exact, since a fair amount of what we use outdoors, such as gardening, ends up in the stormwater network or the environment instead, fewer litres in still means fewer litres to treat, easing pressure across both networks.
Finding the water that goes nowhere
Not all of the savings come from what people choose to use. One property in New Plymouth was found to be losing 60,000 litres a day, roughly what 120 households would use between them, until a smart meter flagged the anomaly and the leak was fixed. That is a single property. Scaled up across a network, it points to a bigger version of the same problem sitting underground.
Taumata Arowai released its first Network Environmental Performance Report in July 2026, giving the country a complete national picture of how our water networks are performing for the first time. The picture it painted was one of ageing infrastructure and years of deferred maintenance finally showing up in the numbers. Using that same dataset, the University of Otago's Public Health Communication Centre calculated a national leakage rate of 22 percent, or around 133 million cubic metres of treated water lost every year, at an estimated cost of $122 million.
Put simply, for every five litres that leaves a treatment plant, one never reaches a tap. Most of that loss stays invisible for a long time. A pipe can leak for years unnoticed, because there is no easy way to see it happening. The damage usually only becomes obvious once a sinkhole opens up or a main bursts, and by then the repair costs far more than it needed to.
We have always had good visibility of how much water enters a network. What has been missing is visibility of where it ends up. Comparing total supply against total consumption confirms a gap exists but says little about where it is or how quickly it is growing. This is where smart meters earn their name. Because they report several times a day, councils can match how much water left the network at a given time against how much arrived at households in that same window and use the difference to pinpoint hotspots. Every leak found this way saves water we would otherwise need to supply, spares infrastructure we would otherwise need to build, and keeps costs off ratepayers' bills.
What a smart meter actually does
At its simplest, a smart water meter transmits usage data back to a council or council-controlled organisation at regular intervals, rather than a manual read every few months. Some existing meters can be fitted with a smart cap that ticks over with the meter and relays the reading, while others are inline meters that measure the water as it passes through. Both achieve the same outcome, regular automated data.
Getting that data home is the harder part. NB-IoT works like 4G, with each meter carrying a SIM card. LoRaWAN uses low-power radio built for small bursts of data over long distances. Christchurch uses its streetlight mesh network, where meters talk to the nearest light and the data hops back across the city. Whichever solution is used, it still has to transmit reliably and consistently from a box buried in the ground. There is no power source down there either, so every meter also needs its own battery.
Installing them at scale is a project in its own right. Christchurch's rollout covers around 115,000 connections, essentially every household in the city, and about 90 percent go smoothly. The crew finds the box, fits the meter, confirms it is talking to the network, and moves on. The remaining 10 percent takes the real effort, with meters buried under driveways, shared connections that need separating, and old installs running under a neighbour's section. Each one needs its own fix, and together they eat up a big chunk of the programme's time and money.
That first round is always the hardest, since everything is being found and sorted for the first time. But it is not a one-off project, as the meters will need replacing when the batteries give out in around 15 years. That first round lays the groundwork, and the work then settles into an ongoing cycle of replacement and upgrades.
A national push, and a chance to collaborate
Smart water metering is picking up pace right across the country. The government has set a five-year window for councils to move away from property-value-based water charging, and from June 2026 the Commerce Commission requires regulated providers to report key performance data, giving the sector a shared baseline for the first time.
Right now, most councils run their own programmes largely in isolation, each choosing its own technology, suppliers and data platforms. A single national solution rarely makes sense given how different every network's budget and priorities are. But the practical challenges look remarkably similar everywhere, from communication reliability and battery life to the awkward installs and the day-to-day work of managing contractors. A crew searching for a meter under thirty years of landscaping in Canterbury is solving much the same problem as one in the Waikato or the Bay of Plenty. The more councils share what they learn, particularly neighbouring ones working in similar conditions, the faster the whole sector moves forward.
Part of a bigger picture
Smart water meters will never solve everything on their own. They sit alongside renewals, capacity planning and long-term investment in ageing networks as one piece of a much larger picture.
But solving a problem starts with understanding it, and that has always been difficult without reliable data. The water sector has spent years making significant decisions with limited visibility into what is actually happening underground. Smart metering will not fix a pipe or build a treatment plant but what it does gives us is the information to make those calls properly, grounded in what is genuinely happening rather than what we assume. That, more than anything, is why it matters.