Apparent losses: meter under-registration, illegal use and data errors

Apparent losses — the World Bank calls them commercial losses — are the most valuable component of non-revenue water per cubic metre, because every litre recovered becomes revenue at full tariff. They are also the hardest to see.

Customer meter under-registration

Mechanical meters slow with age and wear, and almost all meters miss the lowest flows — dripping taps, cistern fills, slow leaks on the customer side. A fleet of 10-year-old 20 mm meters typically under-registers 3–6%; oversized meters on low-consumption customers can miss 10–15%. The best evidence is bench testing a random sample by age and size. In the absence of tests, 2–5% with a ±30–50% uncertainty is a defensible starting point.

Unauthorised consumption

Illegal connections, bypasses and meter tampering are estimated from field investigations: the number of suspected connections multiplied by a typical consumption. For domestic connections use persons per household × litres per person per day; for non-domestic, a litres-per-connection figure. Well-regulated systems have close to zero; informal settlements and intermittent-supply systems may exceed 5% of connections. Carry a ±50% band unless you have survey data.

Data handling and billing errors

Accounts missing from the billing run, estimated reads never corrected, meter-to-account mismatches and corrupt meter-reading practices all reduce billed volume without any water changing hands. A billing audit is the only way to quantify them; 0.2–1% of authorised consumption is a typical range.

Why apparent losses get their own indicators

Apparent losses are reported as a percentage of authorised consumption and as litres per connection per day (not corrected for supply time, because they are driven by customers rather than pressure). Reducing them is usually cheaper and faster than reducing leakage — meter replacement programmes and billing clean-ups pay back in months.