Chief Executive Officer / India
Changes that made unfinished service visible
- Company revenue
- ₹286 Cr
- Year ended March 2026
- Diagnostic centres
- 36
- Six southern Indian cities, June 2026
- Colleagues
- 1,580
- Network workforce at June 2026
Decisions & consequences
01Explaining the visit before collecting the payment
In 2022 we reviewed billing complaints at six centres and found a recurring pattern. The amount shown during booking described a particular service, while patients understood it as the expected cost of the visit. Additional items could be legitimate, but the explanation arrived too late. The finance team could reconcile the account and still leave the family feeling that the original conversation had been incomplete.
I asked operations, finance and the clinical leads to define an itemised estimate for the expected visit. It had to state what was included, which circumstances could change the scope and who would explain a revised amount before collection. Administrative staff were not to infer a different clinical requirement in order to produce a convenient fixed price. Our commercial head worried that a longer estimate would make us appear more expensive than competitors advertising a single number. The centres also expected slower booking calls. I accepted both concerns and authorised a six-centre pilot with additional training time.
Recorded billing complaints fell from 83 among 2,440 visits in the baseline period to 31 among 2,670 visits in the following period. Those were different groups of patients, and the comparison did not isolate the effect of the estimate from every other change. Booking conversations became longer, while fewer families needed a second explanation at the payment desk. We retained the raw complaint counts and reviewed the remaining cases rather than turning the reduction into a claim that every price conversation had been resolved.
The pilot exposed a weakness in our own language. Some staff read the estimate aloud without checking what the patient thought it covered. I would now include that check in the initial training rather than add it after observing the calls. Clear paperwork supports an explanation; it does not establish that the person receiving it has understood the practical consequences.
Reserving time that the appointment count could not see
In 2019 four centres reported recurring delays around visits where patients requested additional assistance. The schedule assumed broadly similar administrative and movement time for every appointment. When a visit took longer, staff tried to recover the timetable by moving the next person through more quickly. That put pressure on employees and left accompanying relatives uncertain about how much of the morning they should reserve.
As chief operating officer, I asked the centre managers and clinical leads to design an appointment-access pilot. Booking staff would ask about practical assistance without trying to assess a medical condition. The centres would reserve suitable time and confirm the arrangement with the patient. Our regional commercial manager objected that the proposal removed 24 bookable slots a day across the four centres. She proposed adding a general warning about possible waits instead. I chose the reserved-capacity approach because a warning did not provide the additional staff time we already knew was needed.
The twelve-week pilot covered 186 visits with agreed assistance arrangements. For those visits, median time from arrival to the start of the booked service was 21 minutes, compared with 39 minutes in an earlier review of 142 comparable requests. The cohorts differed and the measure covered waiting, not clinical quality or the complete duration of a visit. We retained the 24-slot daily capacity reduction and spent ₹18 lakh on staff overlap and practical training. Revenue per available room did not improve during the pilot, which I included in the executive review.
I would change how we first offered the option. Some patients heard the question as an assumption that they could not manage independently. We revised the wording with feedback from those attending and made the choice clearer. The improvement was not simply a longer booking interval. It was an agreed arrangement that respected the person using it and made its operating cost visible.
Finding the owner after the upload had succeeded
A 2024 review identified 127 reports that had been uploaded to the patient portal but had no recorded successful notification. The system team considered the upload complete. Centre teams believed that patients who needed assistance would call. Neither view established that the intended recipient knew a report was available or had an agreed alternative way to obtain it. The clinical communication procedures remained separate and could not be replaced by an administrative delivery count.
I asked our service head to assign one owner to reconcile each unresolved route with the relevant centre, using approved contact and identity-verification processes. The technology lead proposed another automatic notification attempt, which would have been less expensive. I supported a limited manual review first because another message would not resolve an incorrect number or a patient's need for an alternative authorised channel. The medical director made clear that service staff must not discuss the meaning of results. We incorporated that boundary into the work instructions and provided an escalation route for clinical questions.
Within ten working days, 109 cases had a recorded successful notification or an agreed authorised delivery arrangement. Eleven remained open because the intended recipient could not yet be reached, and seven required further identity or authority checks before the route could proceed. We did not count those eighteen as resolved. The review required ₹6.8 lakh of temporary service capacity and exposed weaknesses in how contact preferences were confirmed at registration. It did not establish whether patients had read or acted on the reports, and we did not claim that it did.
I would begin with a clearer definition of each delivery state. Our initial meetings spent too long debating whether a portal event meant that a person had been informed. Once the states were separated, ownership became easier to assign and the technology team could address the recurring causes rather than simply repeat the same notification.