Google and Meta are leading a wave of contract renegotiations that's sending shockwaves through Indian IT. Google cut nearly $50 million from its $200 million annual contract with HCL Technologies, while Meta reduced Wipro's deal by $15-20 million. These aren't isolated incidents—they're part of a broader cost efficiency program where tech giants repurpose budgets for AI-focused spending. Meta also partially pulled the plug on contracts with Accenture, Teleperformance, and Concentrix.
The pattern is clear across sectors. Clients in telecom, SaaS, hi-tech, and financial services are demanding 20-30% discounts on new contracts. This aggressive pricing is eroding the pricing power that Indian IT companies built over decades. Projects are becoming smaller, and clients are increasingly opting for mid-tier alternatives like Persistent Systems, Mphasis, Zensar, and Coforge—which are posting double-digit growth while top-tier players struggle.
Hi-tech companies are the first to renegotiate, and for good reason. They have better visibility into AI use-cases in their workflows and many have built their own AI models in-house. Kumar Rakesh of BNP Paribas explains that these companies understand how AI can reduce costs, giving them leverage to demand better pricing from IT service providers.
This behavior cascades to traditional enterprises through a simple mechanism: cost savings demands magnified by inflationary concerns. Large enterprises across sectors want to reduce IT spending, leading to intense competition among service providers who offer significant pricing discounts to win business. The result? Vendor consolidation and squeezed margins across the board.
Three out of the top five Indian IT services companies reported sequential operating margin declines in Q1FY27. Infosys was the exception, with margins improving 20 basis points sequentially to 21.1%. The others weren't as fortunate. Tata Consultancy Services saw margins decline 130 basis points sequentially to 24%, driven primarily by a 170 basis point impact from wage hikes rolled out in Q1. Wipro's operating margins declined to 16%, down 1.2% year-on-year, due to salary increases, large deal ramp-ups, and ongoing AI investments. Transcripts
The structural problem is stark: wage bills account for 50-60% of overall costs and haven't shrunk in recent years despite AI-driven productivity improvements. On top of this, AI investments in partnerships add another 1-2% to costs. AI resources aren't cheaper either, and significant expenses go toward AI skilling of workforces. Gaurav Vasu of UnearthInsight notes this is causing major margin pressures: despite AI investments increasing and productivity gains being passed on, wage bills remain stubbornly high.
Here's the puzzle: AI is delivering genuine productivity gains, but they're not translating to margin improvements. The reason lies in how these gains are captured. Tata Consultancy Services achieves 10-15% overall productivity when leveraging AI, but this is typically compensated by additional opportunities from customers. When productivity increases, work volume also increases, so total contract value doesn't change significantly. Transcripts
Infosys acknowledges AI deflation in their portfolio but notes they don't quantify it externally. When compression occurs, they typically find opportunities to do more work in other areas, with contract terms and scope getting redefined. The fundamental issue: AI productivity gains are real but not fully captureable by IT service providers due to client market power, competitive dynamics, upfront investment requirements, and the fact that productivity enables more work rather than less spending. Transcripts
While top-tier IT companies struggle, mid-tier players are thriving. Persistent Systems grew 19% year-on-year, Coforge delivered 54.5% growth (partially helped by acquisition), and Mphasis grew 9.2% with its highest-ever quarterly TCV win of $760 million—68% of which was AI-led. These companies posted 20% YoY revenue growth on average, outpacing top-tier firms' 1.5%.
The advantage lies in specialization. Persistent Systems built expertise in product engineering, healthcare, and banking. Coforge developed strong franchises in travel, insurance, and transportation. Mphasis sharpened its focus on financial services. Many smaller firms spent the last decade building modern digital systems rather than maintaining legacy ones, positioning them better for AI-driven transformation projects. As Ramkumar Ramamoorthy of Catalincs puts it: "The future belongs to firms that combine AI with domain expertise. Technology alone is no longer enough".
HCL Technologies CEO C. Vijayakumar estimates AI deflation at 2-3% annually for his company's portfolio. He provided a stark example: "A $100 million deal would be much lesser today—maybe $80 million". That's a 20% reduction in deal value due to AI-driven productivity improvements. Achieving these compressed deal values now requires 25-30% more effort to convert and get to the same number. Transcripts
Brokerages see deeper trouble ahead. Ambit Capital estimated cumulative revenue deflation of 15-20% for the IT industry over the next 3-4 years, translating to 3-4% annual revenue hits. Post-Q1 results, Ashwin Mehta of Ambit Capital expects deflationary pressure to stretch further: "The AI deflation assumption in our strategy note was based on the lower end of the productivity and pricing impact indicated by companies, and the trend appears to be moving in that direction".
BNP Paribas' Kumar Rakesh sees at least 3 percentage point annual revenue growth hit due to AI deflation, continuing for another 3-5 years. Both brokerages arrive at similar annual impact figures, but BNP Paribas projects a longer duration, which could result in larger cumulative effects.
Ambit Capital believes the IT industry hasn't passed the midway point of this deflation cycle for several reasons. First, GenAI adoption and impact are still evolving—the next phase could see greater disruption as AI capabilities improve and expand into additional areas of IT services. Second, meaningful AI deflation impacts are at least 2-3 years away from initial discussions, with the full effect expected in FY27 and onwards. Transcripts
HCL Technologies has quantified AI deflation impacts across service lines: 25-30% productivity benefit in software development lifecycle, 40-50% efficiency gains in business process operations, and up to 75% reduction in personnel in contact centers through Agentic and Conversational AI. As these capabilities mature and enterprises become more sophisticated in implementation, deflationary pressures will deepen. Transcripts
The shift from traditional engineer-based billing to business outcome-driven AI deals is making revenue prediction fundamentally more difficult. Tata Consultancy Services notes that AI revenue is "not like traditional ADM revenue where there is a lot of annuity revenue involved." Many AI projects are one-quarter or two-quarter projects requiring continuous new wins, creating lumpiness in revenue. Infosys acknowledges AI revenue can be "a bit up and down on a quarter-by-quarter basis". Transcripts
The margin profile is transforming too. Wipro explains that large traditional deals involving cost optimization face margin pressures, but net new "Reimagine AI" projects deliver much better margins. Infosys emphasizes they "will not underwrite uneconomic productivity assumptions," preferring not to pursue deals that don't make economic sense. Transcripts
HCL Technologies CEO pushes back against margin compression assumptions: "The shift toward outcome-based engagement is very deliberate. Typically, outcome-based models can deliver higher margins because they allow us to leverage technology and optimization levers more effectively". With over 50% of HCLTech's business already operating under some form of outcome-based model, the transition is well underway.
Indian IT companies face a complex strategic choice. AI-focused outcome-based deals offer higher margin potential, client stickiness, and market differentiation—but come with revenue unpredictability, higher risk, and significant upfront investment requirements. Traditional time-and-material contracts provide revenue predictability and lower risk but face margin pressure, limited upside, and vulnerability to AI automation.
The industry is gradually shifting. Coforge CEO Sudhir Singh captures the sentiment: "In our industry, labour as a default has been disrupted, and that is being replaced by AI-native process redesign and by domain-specialised agents. Firms that continue to bill hours are getting left behind".
For three decades, Indian IT ran on a simple equation: more engineers, more billable hours, more revenue. Generative AI is breaking that equation. The transition period will create revenue volatility and margin pressure, but companies that successfully adapt to outcome-based models will be better positioned for long-term growth in the AI-driven enterprise landscape.
