Home / Zambia / Non-Revenue Water

Zambia metered around 80% of connections — and NRW kept rising. The meters themselves are the suspect.

Zambia did the orthodox thing: a National Non-Revenue Water Management Strategy (2022–2026) and metering coverage close to universal across the eleven Commercial Utilities. Yet losses kept climbing — and audits point at the uncomfortable reason: a large share of installed meters are faulty, under-registering or dead. When the instruments lie, the strategy flies blind. addanode builds the verification layer: independent DMA telemetry, night-flow analysis, meter-fleet health checks and a daily production-vs-billing balance your NWASCO reporting can stand on.

WhatsApp an Engineer Book a Discovery Call
What this solves in Zambia

Three loss problems Zambian CUs recognise.

Metered ≠ measured

A meter that under-registers converts revenue water into "loss" on paper while the customer drinks it for free — and a dead meter billed on estimates cuts the other way. Independent zone inlet metering, reconciled against summed consumer reads, shows which zones' meter fleets are honest and which need the workshop, before another replacement budget is spent on guesswork.

Groundwater-heavy schemes with no balance

The metering gap has a name and a number: a 2022 review of the Auditor General's report found one large commercial utility with only around 70% of commercial properties metered, with a significant share of those meters faulty or non-functional. Lusaka's peri-urban supply leans on boreholes; Copperbelt towns mix sources. Without telemetered source metering, the top line of the water balance is an estimate, and every downstream number inherits the error. Source meters plus tank levels close the balance from the top.

Night losses invisible between readings

Monthly reads can't see the 2–4am inflow that exposes leakage, or the slow overnight tank decay that means a trunk main is bleeding. Telemetered zone meters see both nightly — the strongest, cheapest leak indicators a CU can own.

Both are counted as metered; they distort the loss figure in opposite directions An under-registering meter turns water the customer drank into paper loss, pointing a leak crew at sound pipes. A dead meter billed on an estimate closes the gap on an assumption, so nothing looks wrong at all. Coverage statistics record both as metered, which is why near-universal metering and rising losses are not a contradiction. A meter that under-registers A dead meter billed on an estimate What the customer receives Water, part of which is never billed Water, billed at whatever the estimate says Which way the loss figure moves Up: consumption that was never billed is recorded as loss Either way, depending on how generous the estimate is What the zone then looks like A pipe problem, because the gap is wide Nothing in particular, because the gap closed on an assumption Which crew gets sent Leak detection, to a zone whose pipes are sound None, because nothing looks wrong What the coverage statistic says Metered Metered What actually finds it Zone inlet flow against the sum of consumer reads The same reconciliation, plus a register that never moves
Both count as metered, and they distort the loss figure in opposite directions One turns consumption into paper loss; the other hides the gap behind an estimate. What the customer receives Under-registering Water, part of which is never billed Dead, billed on estimate Water, billed at whatever the estimate says Which way the loss figure moves Under-registering Up: consumption that was never billed is recorded as loss Dead, billed on estimate Either way, depending on how generous the estimate is What the zone then looks like Under-registering A pipe problem, because the gap is wide Dead, billed on estimate Nothing in particular, because the gap closed on an assumption Which crew gets sent Under-registering Leak detection, to a zone whose pipes are sound Dead, billed on estimate None, because nothing looks wrong What the coverage statistic says Under-registering Metered Dead, billed on estimate Metered What actually finds it Under-registering Zone inlet flow against the sum of consumer reads Dead, billed on estimate The same reconciliation, plus a register that never moves
The first problem above, in its two forms. Near-universal metering and rising losses are not a contradiction — a coverage statistic counts installed meters, and both of these columns count as installed.
Local operating constraints

Built for Zambian utility reality.

  • The strategy exists; the instrumentation gap is the blocker. The national NRW strategy names the practices — zoning, pressure management and leakage control. What most CUs lack is continuous data to run them. Our deployments supply exactly that layer, exportable for NWASCO sector reporting and utility boards.
  • Meter verification before meter replacement. Wholesale re-metering is expensive; verification is not. Zone reconciliation ranks where under-registration actually concentrates, so replacement capital goes to the worst fleets first — evidence-led, defensible, phased.
  • Power reality is designed in. After the Kariba drought years, every Zambian operator plans around load management. Metering nodes run on solar/battery and buffer locally, so the night-flow record survives the very outages that stress the network.
  • Existing instruments are read, not replaced. Pulse bulk meters, Modbus and 4–20 mA devices connect as-is; new metering goes only where zones have nothing to read.

Strategy and metering-audit references come from published Zambian sector documents. The pattern is our Southern African utility build, applied to CU structures.

Typical deployment

Verify first, then reduce.

1 · Meter the top

Sources and works outlets telemetered — boreholes included — so production stops being an estimate.

2 · Zone & verify

DMA inlet meters reconcile against summed consumer reads, ranking zones by both loss and meter-fleet honesty.

3 · Hunt by night flow

Night-flow league tables and pressure data route leak crews to physical losses; billing-gap maps route inspectors to commercial ones.

4 · Report & repeat

Recovered volumes documented per intervention — the evidence for NWASCO reporting, tariff cases and the next zone's budget.

First zone typically live in 2–6 weeks on the addaNet platform, beside water telemetry where asset monitoring is wanted too.

NWASCO 2023 on one table

Indicator20222023Reading
Non-revenue water (sector)56.7%57.0%7.23 million m³ more lost than the year before
Cost per service unit vs price—K12.61 vs K7.9The billed 43% is sold below cost
Metering ratio75.5%79%More meters, more losses
O&M cost coverage—all 11 CUs below 100%No utility covers operating cost from tariffs
Best-NRW award—vacantNo utility near the benchmark
Who this serves

Where loss data earns its keep in Zambia.

Commercial Utilities — the eleven CUs whose NRW and metering performance appear in NWASCO's sector report every year · Peri-urban and trust schemes — borehole-based suppliers where the balance starts at the wellhead · Mining towns and estates — Copperbelt and North-Western operations reconciling bulk supply against internal distribution, near our mining condition monitoring work · Industrial water users — processors verifying abstraction and consumption against invoices.

FAQ

Non-revenue water in Zambia — common questions

Why is NRW rising if most connections are metered?

Because installed is not accurate. Audits have found large shares of Zambian consumer meters faulty — under-registering, stuck or bypassed — so billed volumes drift away from delivered volumes even where paper coverage looks complete. Independent zone reconciliation is how you find out which meters to trust, and which zones are leaking physically on top.

Should we replace meters or verify them first?

Verify first, almost always. Zone-level reconciliation costs a fraction of fleet replacement and tells you where under-registration concentrates — often a minority of zones carry most of the drift. Replacement capital then goes where it recovers revenue fastest, with before/after evidence attached.

How does night-flow analysis work on intermittent supply?

In continuously supplied zones, the 2–4am inflow approximates leakage. Where supply is scheduled, the equivalents are charge-window analysis (how fast a zone consumes when charged) and overnight tank decay. All come free once inlets and tanks are telemetered — and none are visible to monthly reading.

Can the records feed NWASCO reporting?

Yes — zone balances, production records and pressure histories export as clean tables, so sector-report submissions and NRW-strategy progress claims trace to continuous measurement rather than year-end estimates.

What does a deployment cost for a CU?

It scales with zones and sources, quoted as a written cost band after a free remote assessment of your network map and existing metering. The worst-zone-first sequence — with recovered water funding the rollout — is the pattern that fits CU budgets.

Does load management break the monitoring?

No — nodes run on solar/battery and buffer locally through outages, syncing when power and connectivity return. The loss record stays continuous precisely through the events that would hole a manual log.

The regulator’s own numbers: NWASCO’s latest published sector report puts sector NRW at 57% — and shows the billed water selling below cost. Our close reading of the report sets out why metering rose while losses rose anyway.

Primary sources

Comparing markets? Non-revenue water in Africa puts the published figures side by side, with what each regulator asks a utility to prove rather than simply to state. It includes the Zambia row, set beside the markets around it.

Find out what your meters are really telling you.

Send us your scheme layout and metering coverage. We'll propose a verification-first deployment that shows where the water — and the revenue — actually goes.

WhatsApp an Engineer Book a Discovery Call