How Indian IT Corporations Have Survived Each Tech Disruption. Why AI Will Be No Totally different

Introduction

The worry is actual.

During the last 12 months or so, each time I learn a put up on the IT sector, it talks virtually about the identical factor.

Each second LinkedIn put up, each enterprise information headline, and each analyst report appears to be signalling the identical factor. Synthetic Intelligence (AI) goes to essentially destroy the Indian IT providers business as we all know it.

Some individuals are saying TCS, Infosys, Wipro, and their friends are heading for a sluggish, painful decline.

That may be a robust declare, and for buyers like me, it requires motion.

However since my 15+ years of investing expertise, I’ve additionally seen that investing in such occasions, in prime corporations of a distressed sector, provides the very best returns in the long run.

I’ve been following the Indian IT sector for some time now. I believe the worry is being exaggerated, not as a result of AI just isn’t disruptive, however as a result of individuals are forgetting historical past.

This business has already been via a number of waves of know-how disruption.

• Every time, the narrative was comparable.
• Every time, the business discovered a strategy to take in the shock and rebuild itself across the new actuality.

What I wish to do on this put up is share with you the historical past. I’ll then examine it with what is occurring right now with AI.

So let’s begin.

The Basis Was Constructed on on-premises all the things

To know why AI feels so threatening right now, you first want to grasp what Indian IT corporations had been truly promoting for the first 20 years of their existence.

Within the Nineties and early 2000s,

Enterprise know-how regarded very totally different from what we see right now.

Massive firms in the USA, the UK, and throughout Europe ran most of their essential enterprise techniques on their personal infrastructure.

Each main financial institution, insurance coverage firm, telecom operator, and manufacturing agency maintained bodily servers on their premises, managed their very own databases, and ran their very own ERP techniques.

They used to maintain a whole IT division simply to maintain their IT infrastructure working.

This created a scenario the place enterprise software program needed to be bought as a product, put in bodily, customised to suit the corporate’s particular workflows, after which maintained indefinitely.

SAP, Oracle, PeopleSoft, Siebel, these had been the platforms that ran international enterprises.

Deploying them was sophisticated, time-consuming, and costly.

American and European corporations shortly realised that sending this work offshore to India was far more cost effective than hiring domestically.

That’s precisely the place corporations like TCS, Infosys, Wipro, and Tech Mahindra stepped in.

They constructed their total enterprise mannequin on being the trusted, reasonably priced, technically succesful associate for precisely this type of work. They provided their clients the next as their enterprise providing:

• Implementation,
• Customisation,
• Utility upkeep,
• Infrastructure administration,
• Testing,
• Manufacturing help, and
• Offshore growth.

The economics for our IT corporations had been easy.

These corporations are billed by the hour and by headcount.

The bigger the group, the bigger the contract.

• Infosys alone went from roughly 1,000 workers in 1993 to over 75,000 by 2007.
• TCS crossed 100,000 workers across the identical time.

The Indian IT business grew right into a $50 billion plus sector based mostly on this mannequin.

The mannequin labored as a result of enterprises genuinely wanted giant, sustained human effort to maintain their advanced on-premise know-how ecosystems operating.

Instance: Infosys is working for Basic Electrical.

Within the pre-SaaS period, Infosys groups helped set up, customise, keep, and help software program operating inside GE’s personal knowledge facilities and workplace infrastructure via long-term manpower-based IT service contracts. The kind of system helped Basic Electrical handle its enterprise operations like finance, worker data, stock, manufacturing, inner workflows, and enterprise knowledge throughout totally different places of work and factories. These are what we used to name ERP (Enterprise Useful resource Planning) techniques provided by corporations like SAP, Oracle, PeopleSoft, Baan, and so forth.

Then SaaS Arrived

Individuals stated the identical issues they’re saying now about AI.

Round 2007 to 2010, one thing began shifting quietly within the enterprise know-how world.

Salesforce had already been constructing its cloud-based CRM platform since 1999. But it surely was solely within the late 2000s that cloud and SaaS genuinely began threatening the on-premise mannequin.

Amazon Net Providers launched in 2006 and commenced displaying enterprises that they didn’t have to personal bodily servers anymore.

Workday was based in 2005 particularly to exchange on-premise HR and finance techniques with a cloud-delivered different.

ServiceNow, SuccessFactors, and later Microsoft’s Azure and Workplace 365 all adopted the identical logic.

The worry on the time was similar to what we’re listening to about AI right now. If enterprises now not wanted to put in and keep software program on their very own infrastructure, what would occur to all these implementation and upkeep contracts?

• If Workday changed PeopleSoft, you wouldn’t want armies of PeopleSoft consultants anymore.
• If Salesforce changed on-premise CRM, these customisation and help initiatives would shrink.

The argument was that SaaS would standardise and commoditise enterprise software program, dramatically lowering the providers alternative.

And to be truthful, this worry was not irrational.

The enterprise mannequin actually was below strain. Among the largest contracts in Indian IT historical past had been long-term utility upkeep and infrastructure administration offers. This was exactly the form of work that cloud and SaaS had been making much less crucial.

Instance: Deloitte is implementing Workday for Walmart.

Within the SaaS period, Deloitte helped Walmart migrate HR techniques to the Workday cloud. They used to configure workflows, combine enterprise techniques, prepare workers, and supply ongoing help as an alternative of managing Walmart’s personal servers. The system helped Walmart handle employee-related operations like hiring, payroll, attendance, efficiency monitoring, scheduling, and HR workflows via a cloud-based platform.

What Really Occurred Is the Extra Vital Story

When individuals unfold a story that the Indian IT Sector could die a sluggish dying, I believe they persistently miss a degree.

Know-how transitions doesn’t remove the necessity for providers.

Sure, they’ll remodel what these providers appear like, however the want for IT providers shall be there.

When a mid-sized American insurance coverage firm determined to maneuver from an on-premise HR platform to Workday, it didn’t simply flip a change.

• It had fifteen years of worker knowledge sitting in a legacy system.
• It had customized payroll guidelines constructed round native tax rules throughout twelve states.
• It had integrations connecting HR knowledge to finance techniques, compliance techniques, and third-party advantages suppliers.
• Its HR group had workflows constructed across the previous system that now wanted to be redesigned totally.

None of that complexity disappeared simply because SAP (ERP) grew to become out of date and Workday (SaaS) was the brand new tech.

When ERP died, and SaaS got here, our consulting and IT providers companies constructed total practices round SaaS implementation.

Deloitte, Accenture, Infosys, Wipro, and Capgemini all of them developed devoted Workday, Salesforce, and ServiceNow practices.

• Infosys constructed a serious Salesforce observe and have become one of many largest Salesforce implementation companions on this planet.
• TCS constructed cloud migration competency.
• Wipro invested in AWS and Azure partnerships aggressively.

The income combine modified. The abilities required modified. The venture profiles modified. However the total demand for expert know-how providers didn’t collapse. It has solely advanced extra.

By 2020, the Indian IT business had grown to over $190 billion in annual income.

It was considerably bigger than it was earlier than the SaaS disruption started.

That’s the quantity individuals overlook after they predict collapse.

Why AI Feels Totally different — And Why That Feeling Is Partly Justified

AI just isn’t SaaS. I’ll not deny that.

So let’s begin with this premise.

AI is genuinely totally different from earlier know-how transitions in a single necessary method.

Earlier disruptions (Y2K, the web, SaaS, cloud) modified the place software program ran and the way it was delivered.

They didn’t change the elemental nature of the human work required to construct, implement, and keep that software program.

AI modifications that.

How the work is completed right now (Pre-AI Period):

• A developer right now makes use of GitHub Copilot to jot down code,
• A testing group makes use of AI instruments to auto-generate check circumstances,
• A help group makes use of AI brokers to resolve first and second-level tickets mechanically.

The precise information work that Indian IT corporations billed for is being compressed.

Not eradicated, however compressed.

A group that beforehand wanted 80 engineers to ship a venture would possibly now be capable to ship the identical output with 45 engineers utilizing AI instruments.

That has direct implications for income per worker, for headcount progress, and for the standard outsourcing economics that made Indian IT corporations so worthwhile.

Nasscom and varied business analysts have flagged this concern explicitly.

Infosys CEO Salil Parekh acknowledged in current earnings calls that AI productiveness instruments are altering how work will get completed internally.

TCS has spoken about reskilling programmes at scale.

These are usually not small operational changes. They’re indicators that the business is aware of its value and supply mannequin is below real strain.

So sure, the AI disruption is deeper in a particular method. It touches the unit economics of the enterprise extra immediately than SaaS did.

Story From The Demand Aspect of AI

Right here is the place I believe the collapse narrative goes fallacious.

AI is not only disrupting IT providers; it is usually creating huge new demand for know-how work.

And far of that demand will stream via the identical Indian IT corporations that individuals are predicting will decline.

Take into consideration what giant enterprises (Consumer facet) truly have to do proper now. Most of them are nowhere close to being AI-ready.

• Their knowledge is sitting in fragmented legacy techniques.
• Their infrastructure was not designed for AI workloads.
• Their governance frameworks haven’t any idea of AI danger or mannequin accountability.
• Their workforce has no coaching in AI-assisted workflows.
• Their safety posture has not been up to date to deal with the dangers that AI techniques introduce.

Fixing all of this requires huge quantities of expert know-how work.

They may want the next:

• Knowledge modernisation,
• Cloud infrastructure upgrades,
• AI mannequin integration,
• Customized AI agent growth,
• Enterprise AI governance frameworks,
• Coaching, and
• long-term AI operations help.

These are usually not easy consulting engagements.

These are massive transformation programmes price a whole lot of thousands and thousands of {dollars} for giant enterprises.

And which corporations have the worldwide supply scale, deep enterprise relationships, compliance experience, and the power to handle advanced know-how transitions at scale?

You guessed it, the identical Indian IT corporations individuals are writing off.

Gartner estimated that international AI spending throughout enterprises will attain properly over $500 billion by 2027. A big share of that can stream into implementation, integration, and transformation providers.

That’s the new SaaS providers financial system.

The one distinction is, this time it’s constructed round AI platforms (like OpenAI, Anthropic Claude, Google Gemini, and so forth) as an alternative of cloud SaaS platforms (Salesforce, Workday, ServiceNow, and so forth).

The Sample Is Constant — However Execution Velocity Issues Extra This Time

After I take a look at the present AI transition and examine it with the SaaS transition, the underlying sample is nearly an identical.

Within the SaaS period, platforms like Salesforce, Workday, and ServiceNow themselves used to seize the product worth.

However the providers layer round these platforms, like implementation, integration, migration, customisation, and managed providers, used to go to IT providers corporations.

Indian IT companies ultimately discovered that technique and constructed very giant practices round it.

Within the AI period, platforms like OpenAI, Anthropic, Google Gemini, Microsoft Azure AI, AWS Bedrock are capturing the foundational mannequin worth.

However enterprise AI adoption requires a providers layer on prime of these platforms. That’s the alternative for IT providers corporations.

However there’s a distinction between SaaS and AI Transition:

What does it imply?

Corporations that transfer slowly shall be left behind extra shortly than they had been within the SaaS period.

That is the rationale why I believe on this AI period, new leaders will emerge. It’s not that TCS, Infosys, and Wipro would go away; in the event that they undertake the AI wave sooner, they are going to proceed to dominate the sector for an additional 10 years no less than (until a brand new disruption comes).

However I’m undecided if such massive corporations are agile sufficient to mould themselves into this new AI avatar. I hope they do, however I’m undecided. That is the rationale why I’ve shifted all my IT Sector holdings to IT Trade Traded Funds.

What Might Really Go Incorrect

Let’s additionally play the satan’s advocate.

The situation that ought to genuinely concern Indian IT corporations just isn’t that AI destroys all service work in a single day. That isn’t how enterprise know-how transitions work.

The true danger is a slower, extra structural margin compression over 5 to seven years as AI instruments cut back the billing hours required for traditional IT work.

Right now, our IT corporations invoice their consumer for the next providers:

• Code growth,
• Testing,
• Documentation,
• Tier-1 help.

If TCS, Infosys, and Wipro proceed to develop headcount at historic charges whereas AI compresses productiveness economics, the income per worker arithmetic begins working in opposition to them.

Their value mannequin, which is essentially constructed on individuals, turns into tougher to maintain if shoppers are paying for fewer hours to get the identical output.

The businesses that can come out of this transition properly are those that genuinely change their supply mannequin.

These corporations will transfer from headcount-driven execution towards outcome-driven and IP-driven supply. They are going to be constructing proprietary AI instruments. They’d create specialised AI practices with skills which are genuinely expert.

That is what used to occur within the early period of IT revolutions. Solely the very best brains are used to get code developer roles. An identical form of developer function (pondering energy) is required once more to construct new and revolutionary AI instruments.

I believe once more, the IISc and IIT stage engineers shall be required to construct AI-based instruments.

TCS, Infosys, and Wipro all have the assets, the consumer relationships, and the worldwide supply infrastructure to make that shift.

Whether or not they transfer quick sufficient is the true query.

Conclusion

Each main know-how transition within the final thirty years has produced the identical worry cycle.

Individuals take a look at what’s altering and conclude that what existed earlier than will cease mattering.

Generally they’re proper about particular roles or particular enterprise fashions changing into out of date.

However they’re virtually at all times fallacious about total industries disappearing.

The Indian IT sector just isn’t a inflexible, single-product enterprise.

It’s a giant, adaptive providers business that has proven (a number of occasions) that it could reshape itself round no matter enterprises want most at a given cut-off date.

However the business will evolve, and there shall be ache.

The management can even change throughout the IT Trade.

That’s what occurred after SaaS. I’ve affordable confidence it should occur once more.

Have a cheerful investing.