In 2019, while I was operations director, I supported a replacement component proposal intended to ease a production constraint. The nominal dimension matched the purchasing description, and the commercial discussion treated the substitution as a straightforward way to maintain assembly. I approved movement before the intended interface had received the technical acceptance it required. That was my decision, rather than an error I could place entirely with a buyer or supplier.

Maya Tershuvani's technical review identified a different fixing interface. Our affected stock comprised 2,400 component sets: 1,800 were still unfitted and 600 were already in completed chairs held in our warehouse. None of those 600 chairs had been delivered. The production team wanted a rapid answer so it could release the waiting orders; purchasing wanted to preserve the benefit of the replacement supply. Maya would not accept a nominal match as evidence of suitability in the assembly.

I withdrew my earlier approval and supported a hold on the full identified stock. The technical team determined the required replacement and acceptance work. Operations organised segregation, rework and the revised delivery discussion. I explained the five-week interruption to the CEO with my own premature approval in the account, rather than presenting technical review as the cause of an otherwise avoidable delay.

Replacement and rework cost ₹3.4 crore. The identified sets were replaced before the held chairs were released through the appropriate process. Those are manufacturing and financial facts within this invented case, not evidence of a clinical outcome or a general product-safety certification. We changed the sequence of purchasing approval so that commercial acceptance could not be mistaken for technical acceptance of the intended use.

The experience affected my later CEO work because an executive can undermine a boundary without explicitly instructing anybody to disregard it. Approving the plan too early can leave a specialist facing a supposedly settled business decision. I wanted future plans to acknowledge unfinished technical acceptance while there was still room to change them.

In 2022 a proposed distribution arrangement offered an attractive route to additional chair sales. The commercial model forecast 4,000 chairs and ₹26 crore revenue over two years. It was still a proposal, with no signed contract or earned revenue. Our sales colleagues had invested in the bid and saw a chance to improve the predictability of demand across the assembly sites.

The problem lay in the requested exclusivity over spare-parts distribution within the proposed territory. Its draft terms would have restricted access for existing members of our 42-partner service network. The new channel wanted to control the complete commercial relationship, not only new-chair sales. Our service colleagues argued that current product users and partners would bear the practical consequences of that condition, although they were absent from the new-sales forecast.

I asked the commercial team to negotiate a reservation for existing service relationships. We separated the chair-sales opportunity from the parts restriction so that the disagreement was explicit. The prospective partner did not accept the revised condition. Some colleagues favoured proceeding and resolving service access through individual exceptions later; I did not think a series of hoped-for exceptions was an adequate basis for a restriction we would sign knowingly.

We declined the arrangement. The ₹26 crore remained a forgone forecast, not a revenue loss recorded after trading. ₹31 lakh of bid expense had already been incurred and was not recovered by the decision. We retained our existing parts-access arrangements; we did not claim that this proved faster repairs, protected a quantified amount of future income or established a clinical benefit.

For me, the case broadened the meaning of channel strategy. A new intermediary can alter the conditions under which an older product relationship continues. The chief executive has to put that consequence into the decision before the commercial proposal becomes a contract. We gave up a plausible opportunity and kept its preparation cost visible because the alternative required a commitment we were not willing to make.

A 2023 capital proposal sought ₹12 crore for assembly automation. Its demonstration used a standard chair configuration and showed an orderly sequence that appealed to both production and finance colleagues. The proposal treated that sequence as representative of the business's work. Before taking it forward, I asked for a review of the configurations in a defined sample of 2,400 customer orders.

The sample contained 1,440 standard orders and 960 configured orders. The latter required combinations of components and assembly steps that the proposed sequence did not fully handle. The issue was not that automation was inherently unsuitable; it was that the investment account had not made the remaining work and its resources explicit. The production sponsor argued that the standard volume justified proceeding, while assembly supervisors wanted the configured work assessed before their teams inherited it around the new line.

With Maya and the operating leaders, we revised the question for the board. Instead of seeking approval for the original machine-led scheme, we proposed ₹4.2 crore for fixtures and training addressing a wider range of configurations. The board approved that programme. We accepted that it would not provide the same standard-model automation and kept the original proposal outside the approved capital plan.

The fixtures and training programme was completed in 2024. We did not describe the ₹7.8 crore difference between the two proposals as a realised saving: one programme had not been bought at a discount, and the scopes were different. Nor did we attach a productivity or clinical-outcome improvement without a suitable measured basis. The decision purchased a particular capability and left other possibilities open for a later investment case.

The value of the review was to make our order mix part of the capital discussion. A model can be common without representing every important demand on the factory. As CEO, I needed the board to see what each proposal would accomplish, what it would leave unresolved and why we were choosing that combination of capability, cost and limitation.

Our 42 service partners had received a common monthly retainer of ₹25,000. By 2024, the service team was questioning whether that uniform amount supported the capacity expected in twelve remote locations. Travel and the distribution of work differed from the other thirty locations. The existing commercial arrangement appeared consistent, but consistency of payment did not mean consistency of the conditions in which partners operated.

Faisal asked for a clear statement of the expense and the obligation we would fund. Some commercial colleagues preferred paying only for completed jobs, reasoning that a higher fixed amount weakened cost discipline. Service leaders argued that available people and practical coverage could not always be maintained solely from a sparse stream of individual tasks. I wanted the decision to recognise that difference without assuming that every additional rupee would translate directly into a better result for a wheelchair user.

We increased the monthly retainer for twelve remote partners to ₹50,000 and left thirty at ₹25,000. The old monthly total was ₹10.5 lakh; the revised total was ₹13.5 lakh. The ₹3 lakh monthly increase meant ₹36 lakh more annual expense. We described the expected capacity and its review in the partner arrangements, rather than allowing the increase to become an unspecified payment for goodwill.

The change was implemented. Its financial consequence was known, but this case does not claim a measured reduction in repair time or an improvement in health. Those would require their own evidence and appropriate professional interpretation. We also did not infer that all remote locations have the same needs, or that retainers are the right mechanism for every service network.

My decision was to put the cost of our network expectation into the operating plan. A chief executive can request broad coverage while approving economics that make it fragile. I wanted our own contribution to that tension examined before treating a partner's difficulty as simply a performance failure. The higher expense remained visible when we discussed the business's margin.

In 2025 I initially supported a proposed total dividend of ₹4.8 crore. Employee shareholders together owned 65% of Zavqumi, including my personal 6% holding, while an external patient-capital investor held 35%. The proposed distribution mattered to people who worked in the company as well as to the outside investor. I shared that interest, and it needed to be acknowledged when I later recommended changing the proposal.

Faisal's review identified a separately assessed ₹3.6 crore requirement for parts working capital. This was funding tied up in the service-support cycle, not a new expense to be subtracted automatically from operating profit. The board needed to consider that use of cash alongside other commitments and the distribution. Some employee shareholders wanted the original payment retained, while operating colleagues wanted more certainty around the parts plan. Neither preference removed the need to understand the cash account.

I revised my recommendation to a ₹1.2 crore total dividend. We followed the applicable corporate approval process for that distribution and separately funded the ₹3.6 crore parts working-capital plan. The reduction from the original proposed dividend was also ₹3.6 crore, but I did not present this arithmetic as proof that one isolated decision explained every movement in the company's cash. Finance continued to account for the wider cycle.

The final distribution and working-capital funding took place. Paying less than originally proposed did not create profit. Holding parts did not mean they had all been used, and the funding decision did not by itself demonstrate a repair or care outcome. I explained those limits alongside the practical purpose of the plan, including the lower distribution attributable to my own holding.

The case tested whether employee ownership would receive a complete financial discussion when the preferred answer changed. I wanted neither a promise of effortless returns nor an appeal to sacrifice that bypassed scrutiny. A credible executive recommendation had to state the alternative use, the personal interests involved and the approvals through which the company would decide.