A smaller electricity bill can leave the operating question unanswered

## Preserve the observation while questioning its explanation

I once made an operating result sound more conclusive than it was. At Vezquri Utility Services, electricity consumption fell from 1.2 GWh in one period to 1.05 GWh in another. The reduction was 0.15 GWh, or 12.5% of the earlier total. I presented that difference as evidence of an efficiency improvement. The customers' production had changed as well, and our account did not establish how much of the difference came from the operating work I wanted to recognise. The arithmetic did not need correction. My explanation did.

That distinction matters because a challenge to an interpretation can easily be heard as a challenge to the underlying records. Colleagues had collected the figures carefully. They were understandably reluctant to see their work described as unreliable when the weakness lay in the claim attached to it. I kept the measured totals in the report and withdrew the stronger conclusion. We could say that electricity use was lower in the later period. We could not yet say that our changes had produced an established efficiency gain of the same size. Preserving both statements helped us avoid defending the wrong part of the account.

The commercial sponsor had a reasonable concern. Customers understood a lower total, and the company needed to explain why operating investment was worthwhile. A qualified comparison might take longer and produce a less convenient answer. Finance also wanted to know whether withdrawing the headline meant that the original programme had failed. I did not believe the available evidence settled that question either. An unsupported success claim does not become proof of failure merely because it is withdrawn. The honest position was that the contribution of the programme still required examination, with the cost and uncertainty of that examination visible.

Engineering head Sonal Qorzethi challenged the interpretation, while qualified specialists retained responsibility for the technical assessment. As COO, I could provide the time, resources and management support for a better comparison. I could not define a method around the conclusion I preferred. We authorised a ₹36 lakh measurement and comparison review. That amount was an expense of understanding the result, not an efficiency benefit or a deduction from a savings total. I wanted the review to remain useful even if its eventual findings narrowed the case for the operating changes we had made.

## Ask what service the comparison represents

An industrial utility is supplying a service under particular conditions. A consumption total has to be understood in relation to that service before it can support a broader performance judgement. Customer activity may differ between periods. The timing and pattern of demand may matter as well as its aggregate quantity. An apparently simple ratio can conceal changes in the work represented by its numerator and denominator. I would ask technical colleagues which conditions are relevant and how they propose to examine them, without assuming that an executive's preferred unit of comparison is sufficient for every operating question.

There is a temptation to solve the problem by dividing electricity use by a production figure and treating the resulting number as definitive. Such a comparison may be informative, but its meaning depends on what the production figure captures and what else changed. Different products, operating patterns or service requirements can complicate a single aggregate account. I would not turn that observation into a universal technical prescription. The practical management question is whether competent specialists have explained the comparison's scope, assumptions and limitations, and whether the report preserves those qualifications when it moves beyond the engineering discussion.

The authority to operate also needs to remain clear. Our company supplies industrial compressed air; it does not manufacture the equipment or generate electricity. Customer installations and customer production decisions sit with their own responsible owners. A desire to understand the service should lead to properly agreed information and access arrangements, not an assumption that our teams may alter someone else's system. The COO's role includes making cooperation possible while preserving the permissions and technical responsibilities that give it legitimacy. A useful explanation obtained through an inappropriate intervention would create a different problem rather than resolve the original one.

Financial and environmental conclusions require further care. The 0.15 GWh difference was not converted into a cash saving or an emissions reduction in this case. Neither conclusion follows automatically from a raw comparison whose cause remains unsettled. Relevant cost arrangements, boundaries and assumptions would have to be examined for the claim being made. I wanted the business to resist carrying one attractive observation through several successive labels until it appeared to prove commercial, operating and environmental benefits at once. Each claim should be supported on its own terms, with the people qualified to assess it involved in the decision.

## Let the unresolved result change the management conversation

A management meeting often wants a clear result because it has to decide what to do next. Leaving a question open can feel like withholding the answer needed for that decision. I have found it more useful to state what can be decided now and what depends on the unfinished work. We could fund the review, retain the original observations and stop using the unsupported headline. We could not honestly close the programme's evaluation with a new success percentage. That was still a management outcome: resources and public language changed in response to a limit in the evidence.

The correction also changed the discussion about accountability. I had endorsed the interpretation, so it would have been unfair to describe the problem solely as a technical reporting failure. The pressure for a concise positive result came partly from leadership. Recognising that did not remove the need for careful measurement or review. It made the organisation's response more credible, because the person who had benefited from the stronger story accepted responsibility for narrowing it. Colleagues could challenge the conclusion without being asked to pretend that the underlying records were worthless or that the original work had no possible merit.

For a board, I would translate the experience into questions about the route from observation to enterprise claim. What changed in the period being compared? Who has examined the explanation? Which part of the reported benefit is measured, inferred or still under review? How does management show the cost of resolving uncertainty? These are oversight questions rather than instructions to adjust equipment or choose an engineering model. Directors should be able to understand why a conclusion is supportable while leaving technical methods and operating execution with the appointed specialists and management team.

I would be cautious about making this into an argument against concise reporting. A board cannot examine every operating record, and a useful summary has to select. The requirement is that selection should preserve the distinctions on which the decision depends. A short report can say that consumption fell, production conditions changed and the efficiency contribution remains under review. That sentence may be less celebratory than a percentage presented as success. It gives the reader a more accurate basis for deciding whether to expand a programme, request further work or change the claim made to customers.

My present test is whether the explanation could survive a result I did not hope to see. If the comparison is designed only to confirm an attractive interpretation, it will be hard to learn from an unfavourable finding. If it begins with the service actually supplied and the conditions that changed, it can improve a decision even when the answer is narrower than expected. The lower electricity total was real. The leadership work was to stop asking it to answer a question it had not yet been equipped to resolve.