Let the invoice change when the method improves

A service company can say it values productivity while earning more when a task takes longer. The contradiction is often manageable for a while: effort varies, clients understand that difficult work takes time, and an hourly model is familiar. Automation makes the tension harder to ignore. If repeat execution becomes much faster, the supplier must decide whether to pass through every reduction, defend an unchanged invoice or renegotiate the proposition. I prefer to make that choice explicit before the old model encourages anyone to conceal what has improved.

Show what became faster and what did not disappear

One of our recurring regression engagements had typically required twelve hundred execution hours per release window. At ₹800 an hour, that produced a ₹9.6 lakh bill. Automation reduced repeat execution effort to four hundred and twenty hours. The reduction was substantial and belonged in the conversation with the client. It did not, however, mean that the whole service had become a four-hundred-and-twenty-hour obligation with nothing else to support.

The suite still needed maintenance as the application changed. Environments had to be prepared, unexpected results investigated and specialist review capacity available during the agreed window. Some of those activities had been poorly distinguished in the earlier execution-led estimate. The automation work exposed that weakness in the contract as well as an improvement in delivery. We needed to describe the continuing service more clearly instead of treating the old hourly total as a permanent entitlement.

The account manager proposed retaining the familiar estimate because the client was accustomed to the amount. I understood the commercial anxiety but rejected that basis. An estimate that no longer describes the work should not be protected merely because changing it is inconvenient. It would also give delivery colleagues a poor signal: improving the method would be celebrated internally while its practical effect was hidden from the customer. That is not a durable way to encourage better engineering.

The delivery head separated repeat execution from maintenance, preparation, investigation and reserved review. Finance then examined the proposed cost and margin assumptions. I wanted both disciplines involved. Technical enthusiasm alone could underprice continuing work, while a purely commercial defence could preserve an invoice without explaining its basis. A credible offer required an account of the obligations and an economic model that could support them over repeated release windows.

Negotiate the obligation rather than defend the old bill

We proposed a defined-scope service at ₹7.8 lakh per release window. It included agreed regression coverage, a maintenance allowance and reserved specialist review capacity. Changes beyond the specified application and coverage assumptions would require a separate discussion. The price was ₹1.8 lakh below the earlier typical bill. It did not represent a promise to absorb every future change or an assertion that the service had become effortless because parts of execution were automated.

The client initially argued that four hundred and twenty hours multiplied by the old rate should determine the whole charge. That was a reasonable starting position under the familiar model. We had to explain why the continuing obligation included activities and availability that the old execution-led estimate had obscured. The negotiation was not improved by pretending the client should simply value our expertise more. We needed to show what we would do, what capacity we would reserve and which changes would alter the agreement.

The client accepted the defined-scope offer after discussion. Its bill for the agreed window fell, and we retained a funded basis for the service. The result was useful, but it should not be turned into a universal claim that fixed pricing always benefits both sides. We accepted more estimating responsibility, and the client accepted boundaries around scope. A different engagement, less predictable change or a different demand pattern could require another model. The point was to choose consciously rather than preserve an inherited contradiction.

The price reduction also had to remain visible in our own reporting. It would have been misleading to celebrate automation while keeping a revenue plan that assumed the old bill continued unchanged. Commercial, delivery and finance colleagues revised the account forecast together. A productivity improvement can strengthen a relationship and still reduce near-term revenue from a particular activity. Leaders need to be able to describe both effects without forcing every good operational decision into an immediate growth story.

Keep the new model open to examination

Defined scope is useful only if both parties can understand and manage it. A contract full of obscure exclusions can reproduce the same distrust in another form. We identified the application coverage, expected release windows and maintenance assumptions in language the responsible client team could examine. We also agreed how a material change would enter discussion. The aim was not to generate a new fee for every small variation, but to avoid an unlimited obligation hidden behind a reassuring fixed amount.

We continued reviewing the work after the agreement. If maintenance demand changed materially, the economics would need attention. If the reserved capacity was rarely needed or the method improved further, that too could become part of a future commercial discussion. A new model should not become another untouchable convention merely because it solved an earlier problem. The company needs evidence about the service it actually provides, not only a persuasive account prepared at the moment of sale.

This is also a people question. Teams notice whether management benefits from their improvements or treats them as threats to utilisation. A company that measures success only through chargeable hours can make better methods feel personally risky. Changing the commercial model does not resolve every incentive, but it creates room to recognise outcomes and capability without asking people to preserve unnecessary effort. That alignment matters if the organisation expects technical colleagues to keep improving the work.

For a board, I would frame the question around the business model: what happens to our economics when our method becomes substantially better? If the answer depends on concealing savings, retaining obsolete effort estimates or discouraging improvement, the strategy deserves scrutiny. Directors should ask management how continuing obligations are priced and how the revenue plan responds to reduced effort. They need not choose the price of an individual engagement to understand whether the model rewards the behaviour the company says it wants.

I do not expect every productivity gain to produce a simple proportional price cut. Services include knowledge, availability and responsibility as well as hours. But those elements should be explained openly enough for a client to decide what it is buying. The invoice can change without the supplier surrendering the right to earn a return on a useful service. The work of leadership is to make the new basis credible, accept the trade-offs and keep examining it as the method and the client's needs continue to change.