Kenya loses 48% of its piped water before it's billed. Here's the data layer that fights back.
WASREB's 18th Impact Report (FY2024/25) put national non-revenue water at 48%, up from 44% — roughly 242 million cubic metres a year produced, pumped and treated, then lost to leaks, theft and metering failures across 94 water service providers. NRW reduction is a programme, not a product — but every working programme runs on the same data foundation: zone metering, night-flow analysis, pressure monitoring and daily input-vs-billing reconciliation. That foundation is what addanode builds.
Three NRW problems every Kenyan WSP recognises.
You can't say which zone loses the water
A single production figure against a single billing total hides everything. District metered areas (DMAs) — bulk meters on each supply zone, read continuously — turn one unanswerable number into a ranked list of zones, each with its own input-vs-billed gap. Reduction programmes start winning the day the worst zone has a name.
Night flow is invisible on manual reads
Between 2am and 4am, legitimate demand is near zero — whatever still flows into a zone is mostly leakage and illegal connections. A meter read monthly by a walker can never see this signal; a telemetered bulk meter sees it every night. Minimum night flow is the single most powerful leak indicator a utility owns, and it's free once the meter is online.
Commercial losses hide behind physical ones
WASREB's framework splits NRW into physical losses (leaks, bursts) and commercial losses (theft, meter under-registration, billing gaps). Zone data separates the two: a zone with high night flow has a pipe problem; a zone with normal night flow but a wide billing gap has a metering or revenue problem. Different problems, different crews, same dashboard.
Built for how Kenyan utilities actually operate.
- Regulator expectations are rising. WASREB has published NRW management standards (developed with JICA) covering pressure management, active leakage control and repair turnaround — and its Impact Report ranks every WSP publicly. Continuous zone data is what turns those standards from audit findings into an operating routine, and exports cleanly for WASREB-style reporting.
- Smart-meter capital is arriving; the zone layer shouldn't wait for it. Nairobi's utility is rolling out ultrasonic smart metering under a KSh 2.57 billion programme, targeting NRW of 39% by 2028/29. Consumer meters take years; DMA bulk metering delivers a loss map in weeks and tells the meter rollout where to go first. (Costing meters? See our vendor-neutral smart water meter price guide for Kenya.)
- Existing meters get read, not replaced. Pulse-output bulk meters, Modbus and 4–20 mA instruments connect as they are. Where zones have no metering, we add ultrasonic or mechanical bulk meters at zone inlets only — the cheapest instrumentation per megalitre of insight.
- Coverage holes and power cuts don't stop the record. Zone meters in low-coverage areas run on private LoRaWAN or 4G; every device buffers through KPLC interruptions so the night-flow record — the very thing you need — never has holes.
The national numbers above are WASREB's, from public sector reporting. The deployment pattern below is drawn from our water-sector work across Southern Africa, adapted to Kenyan schemes.
From one number to a ranked loss map.
1 · Zone it
Define DMAs with your engineers on the existing network map — most schemes zone naturally around tanks and trunk lines. Bulk meters go on zone inlets; no consumer meters needed to start.
2 · Meter & connect
Inlet meters and key pressure points telemetered over 4G or LoRaWAN. Data lands on the addaNet dashboard beside your borehole and reservoir telemetry.
3 · Rank the zones
Night-flow league table plus input-vs-billing gap per zone. The worst zone gets the active-leakage crew first; illegal-connection sweeps go where billing gaps are widest.
4 · Prove the recovery
Re-measure after each intervention. Recovered megalitres become the business case for the next zone, the tariff submission and the board report — evidence, not estimates.
First zone typically live in 2–6 weeks. Hardware pre-configured in Johannesburg; installation by your technicians or local partners under remote guidance, on the addaNet platform.
WASREB Impact 18 on one table
| Indicator | Figure | Reading |
|---|---|---|
| Non-revenue water | 48% (FY2024/25; 43% three years earlier) | About 242 million m³ a year produced but never billed |
| Metering ratio | 97%, unchanged for three years | Meters allocate consumption; they do not find the loss |
| Production vs billed volume | +9.4% vs +2.3% | The gap is leakage, unbilled use and metering error |
| Average utility score | 44% (of 200 points), static | No single-indicator winners |
| Nairobi target | 39% by 2028/29 (smart-meter programme) | Zone measurement is what shows whether it is working |
Where NRW data earns its keep in Kenya.
County water service providers — the 94 WASREB-regulated WSPs whose NRW percentage is published every year, and whose tariff cases improve with metered evidence · Bulk water suppliers and water projects — cross-county schemes billing downstream utilities on volumes both sides can verify · Private and estate suppliers — operators who buy or pump in bulk and resell, where every unmetered branch is margin leaking · Prepaid and kiosk operators — water ATM fleets where the gap between tokens sold and water dispensed is the loss that matters.
Non-revenue water in Kenya — common questions
What is Kenya's current non-revenue water level?
WASREB's 18th Impact Report, covering FY2024/25, put national NRW at 48% — up from 44% the previous cycle — across the 94 regulated water service providers. That's roughly 242 million cubic metres a year produced but never billed, against the World Bank's reference points of about 15% for developed-country utilities and 35% for developing ones.
Do we need smart meters on every household to reduce NRW?
No — and starting there is the classic sequencing mistake. Zone (DMA) bulk metering gives you the loss map for a fraction of universal metering's cost, usually in weeks. Consumer smart metering then goes first into the zones where commercial losses are proven, which is exactly how a constrained capital programme should be aimed.
How does night-flow analysis find leaks?
At 2–4am, real demand is minimal, so a zone's inflow at that hour approximates its leakage plus unauthorised use. Telemetered inlet meters make this visible nightly and trendable over months — a rising night flow flags a new leak long before it surfaces on a road. Pairing it with pressure data localises the search further.
Can this feed WASREB reporting?
Yes. Zone input volumes, billing reconciliation and pressure records export in clean tabular form, so the numbers in your Impact Report submission trace back to continuous measurements rather than end-of-year estimates. The same records support tariff justifications and performance-improvement plans.
What does an NRW monitoring deployment cost?
It scales with zones, not promises: a first deployment covering one or two DMAs is quoted as a written cost band after a free remote assessment of your network map and existing metering. The recovered water in a typical worst-zone intervention is usually worth multiples of the monitoring that found it.
Is this the same as the water telemetry you offer?
It runs on the same platform, aimed at a different question. Water telemetry watches assets — boreholes, pumps, reservoirs. NRW monitoring watches the network's economics — where water enters, where it's billed, and the gap between. Most utilities end up wanting both on one dashboard, which is the point.
Latest figures: WASREB’s Impact 18 report (FY2024/25) puts national non-revenue water at 48%, up from 44%, while the metering ratio held at 97% for a third year — see our analysis of the regulator’s own tables for what that pairing implies about where the losses sit.
Primary sources
- WASREB — Impact Report Issue No. 18 (FY2024/25), official PDF
- WASREB — Impact Reports Issue No. 18 landing page (June 2026)
- WASREB — Non-Revenue Water Management Standards Vol. 1 Guidelines (June 2022, with JICA), PDF
- WASREB — Non-Revenue Water Management Handbook (June 2022), PDF
- NCWSC — WASREB-approved tariff 2025/26–2028/29, Kenya Gazette Notice No. 2710 (27 Feb 2026), PDF
Name your worst zone within a quarter.
Send us your scheme layout and current NRW estimate. We'll propose a first DMA deployment that turns the percentage into a ranked, fixable list — priced for a county budget.