Water Finance and Management reported that Badger Meter announced on Nov. 3 that it had acquired Melbourne, Australia-based "utility technology and remote monitoring company" 37South. Read the Water Finance and Management report on the acquisition. This article does not assess the deal. For an operator weighing remote monitoring, the more useful question is what a meter reports, how often it reports, and what the system does when a read does not arrive. A pilot can produce a clean dashboard long before anyone has checked those things.
Start With the Register, Not the Screen
A remote read is a register value stamped with a time. A dashboard usually shows values derived from those registers, such as flow rates, daily totals, or trend lines. Before a pilot begins, confirm whether the head-end stores interval reads or only periodic snapshots, whether values are cumulative totals, and what the register resolution is at low flow. The reporting interval matters because a small leak usually appears as a slope in cumulative totals rather than as a spike. A long interval can hide that slope until the loss has run for a long time.
Define the Gap Rule Before the First Alarm
A flat line on a dashboard can mean no water use, or it can mean no signal. Those two conditions carry opposite operational meaning, and a system that treats a missing read as zero will report a false calm. Set a written rule that flags any gap longer than a chosen number of intervals as missing rather than zero. Decide in advance whether estimated reads are permitted, how they are labeled, and whether they can ever feed an alarm. Interpolated values should carry a distinct flag in the record so that a later reviewer can tell a measured read from a filled one.
Treat Night Flow as a Baseline, Not a Constant
Minimum night flow is the usual reference for detecting unreported leakage, but it is not fixed. It shifts with season, irrigation, occupancy, and commercial process load in a district metered area. An alarm threshold set from a few warm weeks may fire constantly in a cool month or stay silent through a real problem. The pilot should collect enough history to cover the seasonal range the system will face, and the alarm threshold should be tied to a named baseline window. Record which window was used, who chose it, and when it will be reviewed.
Check the Clock
Timestamps are easy to overlook and hard to repair later. Confirm whether the meter, the communication module, and the head-end share a time base, and whether stored values are kept in local time or in UTC. Ask how the system handles the daylight saving transitions, when an hour repeats or disappears. Also confirm whether each record is stamped at the moment of measurement or at the moment of receipt. A late-arriving read stamped on receipt will misalign with pressure transients, production meter data, or a field crew's visit log. Any comparison across those sources depends on this answer.
Name the Record That Governs Each Use
Operations, customer billing, and leak investigation often draw on the same meters but do not need the same data rules. An operations alarm can tolerate an estimated read if it is labeled. A billing record usually needs a different standard of completeness and documentation. Name the system of record for each use before the pilot starts, so that a dashboard figure cannot quietly become a billing figure or a regulatory figure because it appeared on a screen. Questions about how a particular record may be used in a dispute are outside the scope of this article and belong with the utility's counsel.
What a Pilot Should Show Before Scale-Up
A useful pilot answers a short list of questions. What share of expected reads arrived on time, and how were the missing ones handled? Do remote reads reconcile with manual reads taken at the same timestamp on a sample of meters, and how large is the difference at low flow? How many alarms were raised, and how many led to a confirmed finding that a field crew could verify? Who receives an alert, what they are expected to do, and how long that takes? If the pilot cannot answer these questions with records, the dashboard has not yet shown that it is measuring what the operator needs.
Vendor ownership changes are a separate review item, with their own contract and data-custody questions. The checks above apply whichever company supplies the hardware and software. They are meant to be written down before the first alarm fires, not reconstructed after an incident.