What non-revenue water actually is
Non-revenue water is the gap between the water a utility puts into the network and the water it bills for. It has three buckets: real losses (physical leaks and burst pipes), apparent losses (illegal connections, meter under-reading, theft and billing errors), and unbilled authorised use (firefighting, flushing). In South Africa the total sits at around 37–47% depending on the source — and at roughly R9.9bn a year, it's not an engineering footnote, it's a budget crisis. Every litre lost is treated, pumped and pressurised at full cost, then earns nothing.
The core problem isn't pipes — it's blindness. Most municipalities can't see where the water goes after it leaves the reservoir. You can't fix, bill or prioritise what you can't measure. Smart metering is, first and foremost, a way to see.
How smart metering finds the losses
- District metering (DMAs). Divide the network into zones and meter the flow into each. Compare water entering a zone against water billed inside it, and the leakage reveals itself zone by zone — so you know where to look before sending a crew.
- Night-flow analysis. At 2–4am legitimate demand is near zero. Whatever is still flowing into a DMA is mostly leakage. Continuous metering makes this minimum-night-flow signal visible and trendable — the single most powerful leak indicator a utility has.
- Remote consumer meters (AMI/AMR). Smart meters that report automatically over LoRaWAN, NB-IoT or similar end estimated billing, expose meters that have stalled or are under-reading, and flag continuous-flow patterns that mean a leak on the customer side.
- Pressure monitoring. Excess pressure drives both leaks and bursts. Monitoring (and managing) pressure by zone cuts the loss rate and extends pipe life.
- Leak and burst alerts. A sudden flow spike or pressure drop becomes an alert in hours, not a problem discovered weeks later on a road that's been flooding quietly.
What it realistically recovers
Smart metering is not a magic tap that closes the whole gap — pipes still have to be repaired and connections regularised. But by directing scarce crews and capital at the highest-loss zones first, smart metering and active leakage management can cut NRW losses by roughly 15–20%. On a utility losing R9.9bn-scale value nationally, even a fraction of that, recovered locally, dwarfs the cost of the metering. The financial case is rarely the obstacle; visibility and prioritisation are.
How to start (without metering the whole city)
- Meter the zones first, not every household. Establish DMAs and bulk-meter the inflow to each. This gives you a loss map fast and cheaply, long before universal smart meters.
- Rank zones by night flow. The DMA with the highest minimum night flow relative to its size is where your water — and money — is going. Start there.
- Investigate and fix the worst zone. Use the data to send crews to the right place, fix the leaks, regularise illegal connections, and re-measure to prove the recovery.
- Roll out consumer smart meters where they pay. Prioritise high-consumption and high-apparent-loss areas for AMI rollout rather than blanket replacement.
- Manage pressure. Add pressure monitoring and management in leaky zones to cut the loss rate while you work through repairs.
- Make it continuous. NRW creeps back. Permanent metering and dashboards turn leak-finding from a once-off project into a standing operational discipline.
Designed for South African networks
Municipal reality here means remote reservoirs, intermittent power and patchy connectivity — so the metering has to suit the site. Low-power radios like LoRaWAN and NB-IoT carry meter and logger data over long distances on small batteries; edge logging buffers readings through load shedding and outages so no night-flow record is lost (the very window you most need); and solar power runs remote loggers off-grid entirely. This is the same resilience discipline we cover in our load-shedding guide and connectivity guide.
At addanode this is our water management solution, delivered for utilities and municipalities on one addaNet platform — district and bulk metering, night-flow and pressure analysis, leak alerting and AMI, with the same data feeding effluent and DWS compliance reporting. Because we build both the hardware and the software in-house and support it locally, we can start you on a zoned pilot that finds real losses fast, then scale — sized for South African infrastructure, power and connectivity.
Frequently asked questions
What is non-revenue water (NRW)?
It's water a utility produces and pays to treat and pump but never earns revenue from — lost to physical leaks, illegal connections, faulty or under-reading meters, theft and unbilled use. In South Africa NRW runs at roughly 37–47% of municipal supply, an estimated R9.9bn a year. Every lost litre carries full treatment and pumping cost and returns nothing.
How does smart metering reduce water loss?
By making the network visible. District metering compares water entering each zone against water billed there to localise leakage; night-flow analysis exposes leaks when legitimate demand is near zero; remote consumer meters end estimated billing and flag under-reading and continuous-flow leaks; and pressure monitoring cuts the loss rate. Crews then fix the right pipes instead of guessing.
How much NRW can smart metering realistically recover?
Around 15–20% of losses, when paired with active leakage management and repairs. It won't close the whole gap on its own — pipes still need fixing — but by directing scarce crews and capital at the highest-loss zones first, the recovered value typically dwarfs the cost of the metering.
Do we have to install smart meters at every property to benefit?
No. Start by metering supply zones (DMAs) and bulk inflows — that gives you a loss map quickly and cheaply. Rank zones by night flow, fix the worst first, then roll out consumer smart meters only where apparent losses and consumption justify them. Zoned first, blanket later.
Will the metering keep working with load shedding and remote sites?
It should, if designed for it. Low-power radios such as LoRaWAN and NB-IoT carry data over long distances on small batteries, edge logging buffers readings through outages so no night-flow record is lost, and solar runs remote loggers off-grid. Resilience has to be designed in for South African networks, not bolted on.