Water Is the Other Loss Number: What Non-Revenue Water Teaches Power DISCOMs

Water utilities and electricity DISCOMs share a problem that rarely appears on a dashboard: a large share of what enters the network is never billed. In water it is called non-revenue water (NRW), and recent material from India and the United States shows the same pattern seen in power loss reduction, where installing meters and sensors is the easy part and turning their data into field action is the hard part.

How big the gap is

A March 2026 white paper from meter maker Metron puts US revenue lost to NRW at about $6.8 billion a year, with roughly 2.7 trillion gallons a year entering systems and never being billed, citing Bluefield Research. The paper also cites S&P Global for the finding that the number of utilities losing 30% or more of their water more than doubled between 2017 and 2021, and the US EPA for the view that three-quarters of NRW is easily recoverable. Because the paper is a vendor document, its case-study savings are self-reported and best read as illustrations rather than benchmarks.

Its breakdown is nonetheless useful. Most NRW is physical leakage, while about 13% comes from commercial causes such as billing errors, slow meters and unauthorised use. The two need different tools: leakage needs pressure management, acoustic sensing and repair crews, while commercial losses need accurate metering, billing data cleanup and enforcement. Power utilities will recognise the split between technical and commercial loss.

The paper also notes that S&P Global Ratings now considers water losses in rating utilities' creditworthiness, which moves NRW from an operations metric to a balance-sheet one.

India: city programmes, uneven numbers

An Elets article published on 8 October 2026 describes smart metering in Indian cities as being at a relatively early stage. It reports that AMRUT 2.0, launched in 2021, carries an investment of ₹2.99 lakh crore, and that by mid-2025 states and union territories had taken up 5,873 projects worth ₹82,222 crore, of which 4,676 worth ₹32,793 crore were completed.

On NRW, the article states that Nagpur reduced losses from roughly 40% to under 29% using AI-enabled leak detection, IoT sensors, acoustic monitoring and robotic pipeline inspection, and that Delhi's NRW is close to 48%. It also reports that Chennai Metrowater is rolling out smart meters to about 100,000 commercial establishments and high-rise residential buildings using NB-IoT and LoRaWAN, while Delhi Jal Board has deployed wireless automatic meter reading in several zones. The article does not cite sources for its NRW figures, so they should be treated as reported claims to be verified against municipal data before use in planning.

The rural dimension

Rural supply is a separate challenge. A summary dated 27 July 2026 reports that functional household tap connections reached 82.09% of rural households, and that the Union Cabinet approved Jal Jeevan Mission 2.0 in March 2026, extending it to December 2028 with an outlay of ₹8.69 lakh crore including a central share of ₹3.59 lakh crore. Schemes are being transferred to Gram Panchayats and Village Water and Sanitation Committees for operation and maintenance, with a service norm of 55 litres per capita per day. This is a secondary source and the official notification should be consulted, but the direction is clear: the programme is shifting from building connections to keeping them working, which is an operations and data problem.

What it means for utilities

The lessons transfer in both directions between water and power.

• Measure before you model. Analytics on leakage are only as good as district-level inflow and customer metering. Without a defensible balance, a loss percentage is an estimate rather than a measurement.
• Separate real from apparent loss. Budgets, crews and vendors differ for each. Reporting a single NRW figure hides which lever to pull.
• Budget for the response, not just the sensor. An alert that no crew can reach within days produces no savings. Leak detection needs repair capacity and contracts sized to alert volumes.
• Make the figure auditable. If loss numbers influence tariffs, grants or credit ratings, the method behind them needs to be documented and repeatable.
• Plan for operations handover. As rural schemes move to local bodies, remote monitoring and simple dashboards matter more than sophisticated analytics.

Electricity DISCOMs managing both smart meter programmes and, in many cities, shared municipal infrastructure planning can reuse communication networks, field-force processes and data platforms across the two services.

Key takeaways

• NRW is a revenue and credit issue as well as an engineering one; rating agencies are beginning to account for it.
• Most losses are physical leaks, but commercial losses need separate metering and billing fixes.
• Reported Indian city results, such as Nagpur's fall from about 40% to under 29%, are promising but lack cited sources in the coverage reviewed.
• Jal Jeevan Mission 2.0 shifts focus towards sustaining service, raising the value of monitoring and local operations capacity.
• Analytics deliver savings only when paired with repair capacity and auditable baselines.

Sources

• Non-Revenue Water: How Utilities Are Solving a $6.8B Problem (Metron)
• Digital Transformation of Urban Water Utilities in India (Elets)
• Jal Jeevan Mission: Rural Tap Water Coverage Jumps to 82.09% in 2026 (Plutus IAS)