What we declined, changed and still had to finish

In 2021 a proposed acquisition offered twelve benefit-administration arrangements for ₹14 crore. I favoured it because acquiring relationships appeared faster than building each one, and I believed our service team could finish the untidy parts after completion. Diligence costing ₹45 lakh established a reconciled administration starting point for eight arrangements but left four unresolved. The issue was not whether those four could ever be understood. It was whether we should accept the responsibility and timetable before knowing what work we were taking on.

Nikhil challenged my description of the remaining work as integration. He asked who would explain the opening record to the customer if the original team was unavailable and which assumptions our colleagues would be expected to treat as settled. Commercial colleagues reasonably pointed to the cost of losing the opportunity after spending on diligence. A narrower purchase was discussed, but the seller's offered terms remained for the combined transaction. I took a recommendation to decline to the board chaired by Leela. Qualified diligence owners retained their assessments; my responsibility was the enterprise decision.

We did not complete the acquisition. The ₹45 lakh remained a real expense, and our organic growth plan had to continue without the proposed relationships. I do not describe the ₹14 crore asking price as a saving: we neither paid it nor obtained the asset it would have purchased. Nor did unreconciled administration records prove fraud, an underfunded benefit obligation or a future financial loss.

The consequence was a slower route to growth with a starting position we could explain. My mistake was assuming that capable operations made an unclear inheritance acceptable. I now ask what must be intelligible before completion and what can reasonably remain integration work, with the people who will carry the service involved before the founder's preferred transaction becomes the company's apparent commitment.

In 2022 our transaction-led pricing made active arrangements easier to explain commercially than quiet ones. Yet an arrangement with few new instructions still required record continuity, customer-team changes and support for authorised enquiries. I had allowed that work to depend too heavily on future activity paying for present attention. We allocated ₹18 lakh to a service-cost study and redesigned the renewal offer around a defined base service plus separately described activity.

Commercial colleagues argued that customers would hear a base fee as payment for inactivity. They preferred to keep the familiar transaction structure and recover the expense through volume. Nikhil's team showed why that answer was uncertain: quieter arrangements could remain quiet while the expectation of continuing support stayed intact. I asked the team to describe the actual work included in the base service and the limits of that offer, not merely spread a desired revenue amount across the customer list.

We presented the revised structure at ten renewals. Six customers accepted and four declined. The six acceptances did not establish a universal willingness to pay, nor did the four refusals show that those customers were unreasonable. They chose against the proposition on offer. The company accepted a commercial consequence rather than leave service colleagues responsible for an expectation whose funding depended on an event they could not control.

The ₹18 lakh study was a programme cost, not a recurring saving or a measure of additional profit. We did not claim that the pricing change improved benefit outcomes, which remained with the relevant authorised owners. My judgement concerned the durability of our own administration business. I would start the cost discussion earlier next time, before customers had learned to regard continuing attention as something that arrived free whenever the transaction count fell. A quieter record should change the work required, not make its existence disappear from the service proposition.

A 2023 procurement sought a combined administration and actuarial-valuation service with an indicated ₹1.8 crore annual bid scope. One supplier relationship would have been convenient for the buyer, and our commercial team wanted to keep the offer together. Tezmavi's records and support work could contribute to a qualified actuary's process, but that did not make the company a valuation provider. I was responsible for deciding whether the broader proposal described a capability we actually held.

The case for bidding broadly was not frivolous. The customer wanted simpler coordination, and declining part of the work could remove us from consideration altogether. We examined whether a separately appointed qualified professional could retain the valuation responsibility while we made the administration scope clear. I rejected wording that would make our company appear to own the professional conclusion merely because the commercial document carried one headline. Qualified legal and actuarial owners needed to determine their own arrangements.

We offered administration separately. The customer awarded a ₹72 lakh annual administration contract; actual revenue in the first financial year was ₹54 lakh because work began partway through that year. These figures describe different periods and measures. The ₹1.8 crore original scope was a proposed bid value, not revenue we had already earned or a loss later recovered. We do not claim that the narrower award caused better actuarial outcomes.

I accepted a smaller commercial position that our operating capability could support. It also required clearer coordination with the separately responsible professional parties instead of pretending that a single proposal had dissolved their boundaries. The lesson for me was to challenge the comfort of a broad label before it became a customer expectation. A CEO can respect a buyer's desire for simplicity while refusing to let the enterprise's promise expand beyond the work it is equipped and authorised to perform.

By 2024 customers had learned that involving me could produce a quick explanation of a difficult history. I had encouraged that pattern because personal responsiveness helped a young company win trust. As Tezmavi grew, it also kept responsibility attached to my availability and made service leaders appear less authoritative than their roles required. We committed ₹46 lakh to a programme in which six leaders would take over ten selected relationships through agreed customer handovers.

Some commercial colleagues feared that customers would interpret the change as the founder withdrawing attention. A few service leaders were also cautious: they could not carry responsibility while I continued answering every difficult question directly. Nikhil asked me to distinguish genuine escalation from my habit of being useful. We agreed that the leaders needed access to the history and room to handle the relationship, with my involvement visible when still required rather than hidden behind a nominal handover.

Seven of the ten relationships completed the agreed transfer by the review date. Three continued to need my participation. The result was neither ten successful handovers nor evidence that the remaining leaders lacked ability. It showed where the organisation and customer relationship had not yet moved sufficiently beyond me. The ₹46 lakh was programme expenditure, not an estimate of founder time saved or a valuation uplift.

The practical consequence was a less flattering but more useful succession account. I had to stop using every intervention as evidence of leadership and examine whether it prevented someone else from carrying the role. Leela could ask about the unfinished three rather than receive a general assurance that succession was in place. My next step was to understand the dependencies within those relationships and address them individually. A company does not become durable when its founder announces confidence in successors; it becomes more durable as customers and colleagues can rely on those successors in the actual work.

In 2025 nine customers reached a renewal and portability discussion. Four renewed and five chose to leave. Our records supported work that had to continue with authorised owners after Tezmavi's contract ended. I wanted a service model in which departure did not become an opportunity to make the customer regret exercising that choice. We spent ₹36 lakh on the five completed exit programmes, with the agreed transfer work and responsibility boundaries described separately from any new provider's decisions.

The commercial objection was understandable: a departing customer no longer generated the same future income, while staff still had substantial work to perform. Some colleagues wanted to make exit support a negotiating instrument in the renewal discussion. Nikhil argued that the resulting uncertainty would also consume operating time and expose our own service promise as conditional on continued purchase. I chose to fund the agreed departures and keep future renewal terms distinct from the work already owed.

All five selected exit programmes completed their agreed administration handovers. That result does not mean we certified the receiving providers, the benefit entitlements or the continuing arrangements. Authorised employers, trustees and professional owners retained their decisions. The four renewals cannot be attributed to our portability approach on this evidence; we had no comparison establishing why those customers stayed. The ₹36 lakh was a cost of the exit programmes, not a retention investment with a demonstrated financial return.

The choice made the boundary of our business clearer. We sell administration that should remain understandable when another organisation takes over. I would now discuss the scope and resources of departure at the beginning of a relationship, while both parties can consider it without the pressure of an imminent exit. A customer staying because the service remains useful is a better proposition to examine than a customer staying because the company has made leaving unnecessarily uncertain.