The choices that changed when assumptions became visible

Deferring a seventh centre after changing the unit of demand

In 2022 I supported a seventh-centre proposal using a four-month sample from the surrounding catchment. The presentation described 1,700 completed visits as though they represented 1,700 potential new users. Reconciliation showed that those visits belonged to 340 distinct people. Some people had attended several times. The activity count was not fabricated, but its interpretation was wrong for the investment question. Nor could we simply turn the 340 into a forecast of people who would choose the new site.

The property team wanted to retain the lease timetable and apply a lower conversion factor to the original total. That would preserve the opportunity to secure the building, but it would not repair the distinction between visits and people. I recommended deferral and commissioned a narrower review of access, travel and willingness to use a different location. Clinical colleagues remained responsible for assessing care requirements. We did not use a commercial sample to infer diagnoses, treatment needs or the appropriate frequency of attendance.

The decision wrote off ₹24 lakh of proposal development and cost us the preferred property's availability. Pema Selqashi challenged why the unit had not been checked before the recommendation reached her. I accepted that responsibility. The strategy team had treated a familiar operational report as a market estimate without establishing whether it answered the same question. We changed the proposal review so its main demand statement identified the population, period and counting unit before financial scenarios were compared.

The lease did not proceed on that evidence. We did not announce a replacement centre or claim the deferred capital as a realised return. The useful outcome was a decision that remained open until the access case could be understood. I learned to distinguish three questions that the original chart had compressed: how much service had been delivered, how many people had used it, and who might use a proposed location. Each can matter, but one count cannot answer all three. The loss of a property was a visible cost; the unsupported certainty in our recommendation was the underlying issue we needed to correct.

Finding twelve transfers inside an eighteen-person expansion plan

In 2024 a proposed centre opening was described as adding eighteen professionals to the network. The staffing schedule contained six external recruits and twelve colleagues proposed for transfer from existing centres. All eighteen roles were intended for the new location, but only six represented additions. My strategy paper repeated the larger claim without explaining what would happen to appointments where the twelve colleagues currently worked.

Operations argued that the transfers could be absorbed through schedule adjustments. Clinical leaders asked for a profession-by-profession review because a general headcount could not establish whether particular services would remain available. We compared the proposed deployment with existing commitments and the skills required at the new site. I rejected the idea of describing overtime or shorter appointments as automatic substitutes for missing professional capacity. Those options required clinical and employment review; they were not assumptions I could insert to make an opening forecast balance.

We deferred ₹3.8 crore of fit-out and spent ₹18 lakh over seven weeks examining a smaller opening option, recruitment timing and the effect on current centres. The investment shareholder preferred to secure the planned launch date. I presented the alternative as a choice about where the network would provide service, not a simple contest between growth and caution. A transferred colleague could improve access in one place while reducing it elsewhere. The board needed to see both consequences before committing.

The revised recommendation separated six proposed recruits from twelve transfers and made any opening conditional on an agreed clinical and operating plan. It did not claim that all six had joined, that every vacancy could be filled on schedule or that additional rooms equalled usable capacity. The deferred fit-out remained an unresolved commitment rather than a booked saving.

This case changed how I present staffing in strategy papers. I ask for additions, departures and internal moves separately, with the relevant professional mix and the effect on existing obligations. A new site's roster is only one view of the decision. The network account may reveal that the same capacity has been counted twice in the story of growth.

Reworking a home-service option around time spent travelling

A 2025 home-service proposal treated a weekly resource envelope of ninety-six paid hours as ninety-six hours available for appointments. The underlying route model already required thirty-six hours for travel, documentation and other paid duties, leaving sixty appointment hours in that defined scenario. The commercial forecast used the larger figure while the operating plan used the smaller one. I had accepted the forecast because home delivery appeared to extend access without another property commitment.

The commercial team proposed increasing the number of visits in each day to preserve expected income. Clinical leaders explained that suitability and appointment duration depended on the person's needs and professional judgement. I did not turn the capacity gap into a standard shorter appointment. Instead, strategy and operations compared a narrower pilot geography, different scheduling assumptions and the option of retaining centre-based delivery where it was appropriate. The comparison kept travel inside paid time and distinguished capacity from the revenue that might eventually be earned.

We spent ₹16 lakh over six weeks revising the pilot design and withdrew the ninety-six-hour appointment assumption. The revised scenario retained sixty appointment hours within the same total resource envelope. It was a planning model, not a promise that every hour would be booked or that every household could be served. The proposed area was narrowed so the team could test actual travel and demand before recommending wider coverage. We also retained a route for clinical leaders to decline home delivery where it was unsuitable.

The outcome was a smaller, more explicit pilot proposal rather than an established improvement in health outcomes or a measured increase in revenue. Naren Dervqeshi asked that the board paper show the resource reconciliation beside the financial comparison. That made the access benefit and the time cost visible in the same decision.

I now ask whether a new channel creates usable capacity or changes where existing capacity is spent. Home service can be valuable, but its attractiveness should survive an account of travel, professional judgement and uncertain utilisation. Removing the building from a plan does not remove the operating constraints that determine what can actually be delivered.