The Indian IT Layoff Cycle: What Is Actually Happening in 2026

The Expectation

The default belief in the Indian IT services industry has been remarkably stable for two decades:

"IT services is cyclical. There are bad quarters, but the model always recovers because global enterprises will always need Indian talent at Indian prices."

This belief is not unreasonable. It was true for a long time. The 2008 financial crisis caused a dip, but by 2010, hiring was back. The 2016-17 automation scare led to restructuring, but headcount recovered. COVID caused a brief freeze, then the 2021-22 hiring boom was the biggest the industry had ever seen.

The pattern trained an entire generation to believe that every downturn is temporary.

The related assumptions are:

• Large IT services companies are "too big to fail" and will always find new revenue streams
• The cost arbitrage advantage of Indian engineers is permanent
• If one technology wave ends, the next one creates equivalent demand (mainframes → client-server → web → cloud → AI)
• Personal job security comes from staying long enough to become a "trusted resource" to the client

Each of these assumptions is being tested simultaneously in 2026, which is why this cycle feels different — because it is different.

The Reality

What is happening in Indian IT services in 2026 is not a cyclical downturn. It is the beginning of a structural repricing of the core business model.

To understand why, you need to understand what the IT services model actually sells. It does not sell technology. It sells labour arbitrage. A TCS or Infosys contract is fundamentally: "We will provide X engineers at Y hourly rate, which is 40-60% less than what you would pay domestically."

That model has three structural problems that are converging simultaneously:

Problem 1: AI is automating the bottom of the pyramid

The IT services revenue pyramid has always been wide at the base. For every architect billing at $80/hour, there were 8-10 engineers billing at $25-35/hour doing maintenance, testing, bug fixes, and support. That base is shrinking.

Not because AI can replace architects — it cannot. But because:

• Automated testing tools (powered by LLMs) are replacing 40-60% of manual QA effort on new contracts
• L1 and L2 support tickets are increasingly handled by AI agents, not human operators
• Code maintenance and bug fixing — the bread and butter of IT services — is being partially automated through AI-assisted development
• Documentation, compliance checking, and basic code reviews are being automated

The key point: clients are not firing their Indian vendors. They are renegotiating contracts with 20-30% fewer headcount requirements for the same scope. The work gets done with fewer people. That is a permanent reduction, not a pause.

Problem 2: The margin squeeze is structural

IT services companies historically operated at 20-25% operating margins by billing engineers at 2-3x their cost-to-company. When headcount requirements drop per contract but the work still needs delivery, the billing model breaks.

Companies are responding by:

• Not replacing attrition (natural headcount reduction of 12-18% per year)
• Eliminating bench periods — if you are not billed within 60-90 days, you face performance management
• Reducing fresher intake (TCS hired 40,000+ freshers in FY22 versus sub-20,000 in FY26)
• Pushing mid-level engineers into "upskilling programs" that are holding patterns, not genuine skill development

Problem 3: The next wave does not need the same headcount

Every previous technology transition — mainframe to client-server, client-server to web, web to cloud — created equivalent or greater demand for human engineers. The cloud migration wave (2018-2023) was a goldmine for IT services.

The AI wave is different. It is the first technology transition where the technology itself reduces the need for human labour in technology delivery. Previous waves were tools for humans. This wave is a partial replacement for humans.

This does not mean IT services will disappear. TCS is not going bankrupt. But a company that employed 600,000 people to deliver a certain volume of work may need 400,000 people to deliver that same volume in 2028. That is 200,000 roles that are not coming back.

What the quarterly numbers actually show

Revenue is flat or slowly growing. Headcount is declining. Revenue per employee is rising. This is the mathematical signature of a model that is producing the same output with fewer people.

Related context: Salary Reality Check, CTC Decoder, more in Career Reality Checks.

Salary and Growth Reality

Cross-check your take-home with the CTC Decoder and compare ranges in Salary Reality.

Where Most People Get Stuck

The most common response to structural change is to wait it out. And in the context of Indian IT services, that response has been reinforced by 20 years of successful waiting.

The 2008 crash recovered. The automation scare passed. COVID was temporary. So this will pass too.

That pattern recognition is not irrational. It is just wrong this time, because the underlying structure has changed.

Why people stay when they should be moving

Sunk cost anchoring: "I have 10 years at Infosys. My gratuity, my internal network, my client relationships — I cannot walk away from that." This is real. Gratuity at 10+ years is meaningful. But it is a one-time payment that is dwarfed by the cumulative salary differential of switching.

The comfort of familiarity: IT services companies are predictable environments. You know the appraisal cycle, the promotion bands, the delivery model. A GCC or product company is unfamiliar territory with different expectations. The unfamiliarity feels like risk, even when the familiar environment is the riskier position.

Skill confidence gap: After years of working in a specific client's ecosystem with specific tools, many professionals genuinely do not know if their skills are transferable. They have not interviewed in 5-7 years. They do not know what a DSA round looks like in 2026. That knowledge gap becomes a barrier to action.

The "one more year" loop: There is always a reason to wait. A pending promotion. A bonus cycle. A visa application. An ongoing project. Each delay is individually rational but collectively devastating. Three "one more years" is three years of compounding opportunity cost.

The structural trap

The deepest trap is that IT services companies are not designed to prepare you for leaving. The skill development, the project allocation, the career pathing — all of it optimizes for the company's billing model, not for your market value. A "senior consultant" at TCS with 10 years of experience may have deep client domain knowledge but shallow technical depth compared to a 5-year engineer at a product company.

Every additional year in a narrowing model makes the transition harder, not easier. The window does not stay open indefinitely.

If this matches your current situation, run the Resignation Risk Analyzer before making your next move.

Who Should Avoid This Path

Not everyone in Indian IT is facing the same situation. Some people are in a structurally different position and this article will not apply to them.

If you are already in a product company or GCC, the dynamics discussed here are largely about the services billing model. Your situation has different risks and different ceilings — but the services-specific structural pressure described below does not apply to you directly.

If you are in your first two years, you have maximum flexibility. The sunk cost problem that traps mid-career professionals has not caught you yet. Your best move is information gathering, not anxiety.

If you are in a genuinely specialized niche — embedded systems, mainframe modernization with deep domain expertise, SAP S/4HANA migration architecture — the generalist squeeze discussed here affects you less. Niche expertise with genuine scarcity still commands premiums, even in a contracting services market.

The people most at risk — and for whom this article matters most — are generalists in a model that is being repriced.