The verdict in three sentences
A water operator with 28,000 customer accounts cannot afford to bill on estimated reads or to leave stuck meters undetected for months. A custom metering and billing platform that links photo-verified reads, tiered tariffs and digital payment cuts commercial losses from the first billing cycle. In Singapore, where non-revenue water targets are among the strictest in the world, the case rests less on recovering lost volume and more on accuracy, auditability and customer self-service.
Where non-revenue water comes from
Non-revenue water combines two families of loss. Physical losses (pipe leaks, tank overflows) belong to operations. Commercial losses (stuck or under-registering meters, estimated reads, unauthorised connections, data entry errors, unbilled accounts) belong to the information system. Singapore's national utility reports non-revenue water around 5 %, while the Antananarivo case in our French edition runs at 30 %. A private operator serving condominiums, industrial estates or a business park in Singapore sits between those extremes, and every point of loss shows up in its margin.
| Loss source | Typical share of non-revenue water | Software lever |
|---|---|---|
| Network and tank leaks | 50 to 65 % | Zone balance, alert on production vs billed gap |
| Stuck or under-registering meters | 10 to 15 % | Detect zero or sharply falling consumption |
| Estimated or wrong reads | 8 to 12 % | Tablet reads with timestamped, geotagged photo |
| Unauthorised connections | 5 to 10 % | Cross-check connection map against accounts |
| Unbilled accounts | 3 to 6 % | Single register of account, meter, supply point |
| Tariff errors | 1 to 3 % | Configured tiered tariff, no manual maths |
What the platform must do, module by module
- Meter reading on tablets or AMR import: route by zone, index entry, mandatory dial photo, geotag, offline mode. Where smart meters exist, readings are imported automatically and only exceptions go to field staff.
- Tiered billing in SGD: block tariffs, fixed charges, water conservation tax and GST at 9 %, all versioned with effective dates so a tariff revision applies on the right day.
- Digital payment: PayNow QR and GIRO direct debit for Singapore customers, card payment on the customer portal, automatic reconciliation.
- Anomaly detection: zero consumption for three cycles, drops above 50 % versus the 12-month average, unexplained spikes, meters never photographed.
- Collections: reminder emails and SMS, payment plans, disconnection and reconnection tracking.
- Dashboards: volume produced per zone, volume billed, non-revenue water rate, collection rate, days sales outstanding.
| Option | Cost, 2026 order of magnitude | Fit for a mid-size private operator |
|---|---|---|
| Spreadsheets and paper | Near zero licence, high hidden cost | No detection, billing delays |
| Enterprise utility CIS | Licence and integrator often above SGD 250,000 | Powerful but heavy for 28,000 accounts |
| Generic billing SaaS | SGD 2 to 5 per account per year | No photo reads, limited tiered tariffs |
| Kolonell, medium application | 4,000,000 to 7,000,000 FCFA (6,100 to 10,670 EUR) | Reads, billing, payments, dashboards |
| Kolonell, complex application | 7,000,000 to 15,000,000+ FCFA (10,670 to 22,870+ EUR) | Adds advanced detection, GIS, customer portal |
| Kolonell maintenance | 35,000 to 120,000 FCFA per month (53 to 183 EUR) | Fixes, minor changes, monitoring |
A smaller first phase focused on reads and billing can fit the simple application bracket (2,500,000 to 4,000,000 FCFA, i.e. 3,810 to 6,100 EUR). Hosting, SMS and payment provider fees are priced in the quote.
A realistic rollout plan
Weeks 1 to 4: clean the customer register and link every meter to a geolocated supply point. Weeks 5 to 10: build the reading app and billing engine. Then run one pilot zone of 3,000 accounts through two full billing cycles before going live everywhere. Plan four to six months in total for 28,000 accounts, with parallel running of the old system during the pilot.
Mini case study
Wei Ling, COO of a private water operator in Singapore serving 28,000 accounts across industrial estates and private estates, distributes 900,000 m³ a month. Non-revenue water sits at 9 %, i.e. 81,000 m³. Assume one third is commercial, 27,000 m³. If photo reads and stuck-meter detection recover half, 13,500 m³ more are billed each month. At an average of SGD 2.80 per m³ (2026 order of magnitude), that is SGD 37,800 a month. Add 40 staff hours a month saved on manual re-keying at SGD 35 an hour, SGD 1,400. A complex-tier build of around 20,000 EUR pays back in well under a quarter.
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
FAQ
Can field staff work without mobile coverage?
Yes, reads and photos are stored offline and synchronised later. In basements and plant rooms, where coverage often drops, this avoids losing 5 to 10 % of a day's reads.
How is a mid-cycle tariff change handled?
Tariffs carry an effective date and the engine prorates by days when a change falls inside a period. That removes the 1 to 3 % of errors typical of manual calculation.
Can customers pay by PayNow and GIRO?
Yes, each invoice carries a unique reference and reconciliation runs automatically. GIRO typically lifts on-time payment above 90 % for enrolled accounts.
How fast does non-revenue water move?
The first anomalies surface after the second read cycle, about 60 days. A reduction of 1 to 3 points in the first year is a reasonable target for a well-run network.
What happens to historical data?
We migrate 12 to 24 months of reads and payments from the existing system or spreadsheets. That history feeds the consumption averages used by anomaly detection.
Let's scope your project. We define the reading, billing and payment modules with you, then a pilot zone you can deploy within a few months. Detailed quote within 48 h. WhatsApp +221 77 596 93 33.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.
