A smaller team augmented by AI tooling can now credibly bid against a workforce many times its size

Reuters reported this week that TCS Chief Executive K Krithivasan told the news agency that roughly 80% of the company’s contracts within its finance, human resources, and other business services segment are now based on outcome performance measures, a proportion that has reportedly doubled since AI went fully mainstream in late 2023.

Cognizant put it plainly: with AI, the fundamentals are shifting, and clients now expect measurable outcomes rather than hours logged. Infosys faces the same pressure. This is not a rounding adjustment to a familiar business model. It is the business model changing underneath the industry that built modern Bengaluru.

The mechanism is straightforward enough. Existing customers, having watched AI compress the effort behind routine service delivery, are demanding a share of the savings before the vendor even proposes them. Some clients are not negotiating the discount at all.

They are moving the work in-house, having concluded that a capability once available only from a large outsourcer can now be assembled internally with considerably less external dependency.

Industry executives described both patterns simultaneously: shrinking fees on retained work, and outright loss of other work, layered on top of contracts that are shorter than before because nobody, client or vendor, is confident enough in the technology’s trajectory to commit to the five-year terms that once defined this industry.

Shorter contracts are themselves worth sitting with. A five-year deal was not merely a revenue commitment. It was an admission that switching costs were high enough, and the technology stable enough, to justify a long relationship.

When contracts compress to 12 or 18 months, both sides are quietly conceding that nobody can confidently price a service five years out, because nobody knows what a unit of that service will cost to deliver by year three.

Meanwhile, the entry-level pipeline that once absorbed hundreds of thousands of engineering graduates a year has narrowed sharply. TCS cut its net headcount by more than 23,000 in the last fiscal year, its fresher intake guidance well below the pace of a few years ago.

Wipro trimmed fresher hiring guidance too, with reports of recruits waiting months for onboarding that may not arrive. I have written in this column before about what disappears when the junior rung is removed: not simply cost, but the years of mistake-making and correction that once produced the senior engineer capable of judging AI-generated output rather than merely accepting it.

That skill gap does not show up on a quarterly call. It shows up three years later, in exactly the client engagements now being priced on outcomes nobody can yet reliably forecast.

Against this, a second story is unfolding among the tier-two firms, and it complicates the usual narrative of AI simply gutting Indian IT. Persistent Systems posted 16.1% year-on-year revenue growth in its June quarter, its 25th consecutive quarter of sequential growth, with record order bookings.

Coforge grew revenue by roughly a third year-on-year, aided by acquisitions but also by genuine competitive wins. Both firms are unmistakably smaller than TCS, Infosys, or Cognizant. Neither can match the bench strength the majors have spent decades building. And increasingly, that bench no longer decides who wins the contract.

This is the detail that should worry the majors most. For years, an enormous employee base was itself a competitive moat: the ability to staff a large, complex engagement quickly, absorb volatility, and offer redundancy a smaller rival could not.

AI erodes that moat directly, because a smaller, more concentrated team augmented by AI tooling can now credibly bid against a workforce many times its size. Scale, which used to be the majors’ answer to every RFP (request for proposal), is turning into overhead that outcome-based pricing punishes rather than rewards.

None of this means the large firms are in terminal decline. TCS, Infosys, and Cognizant retain client relationships, domain depth, and delivery infrastructure built over decades that a Coforge or Persistent cannot replicate overnight, and Cognizant’s own large-deal pipeline this year shows the majors can still win at scale when they choose to compete on outcomes rather than headcount.

But the next several quarters look less like a gentle transition and more like a genuine repricing of the entire industry’s value proposition, conducted in public, one earnings call at a time. Expect margin pressure as outcome-based contracts absorb AI’s productivity gains before shareholders see them.

Expect continued softness in fresher hiring, since the economics of training juniors no longer pencil out the way they once did. Expect further bifurcation, with nimbler tier-two players continuing to take share in specific verticals where a smaller AI-augmented team can move faster than a legacy delivery organisation built for a different era.

The firms that navigate this well will not be the ones with the largest workforce or the most sophisticated AI tooling. They will be the ones that work out, ahead of their clients, which parts of a contract genuinely deserve outcome-based pricing and which are being repriced simply because the word “AI” now appears in every renewal negotiation.

That distinction requires exactly the kind of judgement this column has argued is becoming scarce even as answers become abundant. The next two or three quarters will show which of these firms actually possess it, and which have merely been fortunate that nobody has yet asked the harder question. That question is coming, and it will not wait for a convenient quarter.

Disclaimer: The views expressed are the author’s own and do not reflect the official policy or position of Financial Express.”