Proof

How a saving gets measured.

Every percentage on this site is the difference between two measurements taken with the same instruments on the same circuits. This page is the method, the instrument, and the public data behind the problem.

The measurement method

Four steps · in order

A saving is a difference between two measurements, not between a measurement and a model. Same circuits, same instruments, same resolution on both sides of the install, both periods corrected for degree-days and occupancy before subtracting.

01Baseline before anything is installedTemporary metering on the circuits that matter, at one-minute resolution, over a period long enough to cover the site’s working pattern. No nameplate arithmetic, no estimates from the bill.

02Normalise for weather and occupancyA cooling load compared across two different summers is not a comparison. Consumption is corrected against degree-days and against the site’s own occupancy hours before any figure is quoted.

03Install, then measure the same circuitsThe post-install measurement uses the same instruments on the same circuits at the same resolution. A saving is the difference between two measurements, not between a measurement and a model.

04Report with the method attachedThe written report states the measurement window, the normalisation applied and the residual uncertainty. A figure without those three is not a result.

The instrument

A claim is only as good as the thing that produced it. These are checkable against the standard rather than against us.

Sampling resolution60 s

Current accuracy±1% · IEC 62053-21

Measured channelsV · A · kW · kWh · PF

Temperature input±0.5 °C

Local data retention30 d

The problem, in public numbers

NEPRA · FY2024-25

These are not our figures. They come from the regulator’s own annual report, and they are the reason the work exists.

17.55%Average DISCO T&D lossesAgainst an allowed ceiling of 11.43%.SourceNEPRA State of Industry Report, FY2024-25. Public document.

265bn PKRLost to excessive T&D losses in one yearThe gap between actual and allowed losses.SourceNEPRA State of Industry Report, FY2024-25. Public document.

What we can and cannot do today

Stated plainly, because a procurement team will find this out anyway and it is better said here.

In serviceACMC30 monitoring and relay control, deployed and measured on commercial sites.

In serviceEnergy Audit, delivered as a measured written report.

In serviceMetersBoard, running as a meter data management platform.

PrototypeSmart47 is a lab prototype refined through proof-of-concept work. It is not field-deployed and there is no installed base.

Not yetRemote disconnect through a retrofit needs a relay the conventional meter may not have. Retrofit delivers reads, interval data, tamper events and loss visibility — not switching.

On case studies

We have not published one. A case study here carries five things every time — context, measured baseline, what was deployed, the result, and how it was verified — and no deployment has yet supplied all five.

Nor are there quotes anywhere on this site. A quote carries a named person and a title, or it is cut. An anonymised endorsement is not evidence.

If you want to see the method applied to a building rather than described, a site assessment produces exactly that — for your own facility, with your own numbers.

Book a site assessment

Two engineers, one day on site, a written report with measured baseline and modelled payback.

Book a Site Assessment