How much revenue leaks in a rural water service — and how to measure it
“We’re losing money, but we don’t know how much.” It’s the sentence we hear most. Revenue leakage in a water service isn’t one hole: it’s the sum of small, diffuse losses, each invisible inside its own file. Measuring them starts with naming them.
The four sources you can quantify
- Unmatched payments — collected money that can’t be tied to any subscriber or reading.
- Duplicates — the same transaction recorded twice: revenue that doesn’t really exist.
- Billed-vs-collected gap — by site, what was invoiced but never collected.
- Wrong-site & off-roster collections — payments attached to the wrong water point, or collected by an agent who isn’t on the roster.
Summed by site and by agent, these give an estimated leakage figure, in €.
From gross amount to recoverable
Not all leakage is recoverable. So we apply an explicit recovery assumption — say 60% — to estimate what can realistically be recovered once the anomalies are worked. The number is useful for prioritisation, as long as it’s never presented as guaranteed revenue.
The honesty rule
The recoverable figure applies an assumption, not a promise. An “unmatched” payment can come from a missing file as much as from fraud. Dalili produces decision support — where to look first — not a certification or an audited financial statement. A human decides.
Measuring leakage doesn’t remove it. But you only fix well what you first managed to quantify.