Investment Insight Series
The GenAI Paradigm Shift
in IT Services
A holistic lens on the structural transformation of the technology ecosystem. From first-order productivity gains to the terminal risks facing legacy business models.
Audio Report
A Sector at the Crossroads
The Indian IT sector is not merely an industry; it is the modern economic engine of the nation. Contributing over $250 Billion in annual revenue and employing a direct workforce of over 5.4 Million, it has been the primary vehicle for India's service-led growth.
It absorbs the bulk of the country's engineering talent and anchors a massive ancillary ecosystem—from commercial real estate to urban consumption. With a significant weightage in the Nifty 50, its health dictates broader market sentiment. Any structural threat to this sector will therefore have strong macroeconomic implications for the country and the stock market, making the understanding of this paradigm shift extremely important.
This is because GenAI is not just another technology upgrade—it represents a fundamental discontinuity. Unlike previous cycles that expanded the scope of services, this shift threatens to decouple revenue growth from headcount, challenging the labor-centric foundation upon which the entire industry was built.
Scope of Study
This study is rigorously focused on pure-play IT Services & BPO companies. We have deliberately excluded Engineering R&D (ER&D) and Software Product firms to isolate the specific impact on the "Services" business model.
Why this distinction? Pure-play IT service providers primarily monetize human effort. Their revenue has historically been a linear function of headcount. This makes them uniquely vulnerable to GenAI's ability to automate cognitive tasks.
"I possess neither a crystal ball nor a hotline to the future. This is not a prescriptive report from a high-tower consultant, so if you are looking for absolute certainties, you might want to stop reading here. I am simply a fund manager building a mental model to navigate the chaos—hoping to spot opportunities and dodge potential disasters. Since this turned into quite the manifesto, I’ve added a table of contents above to help you navigate the sections at your convenience."
— Chaitanya Shah, CIO, Amaltas
1. Decoding the Business
Before categorizing the players, we must first understand the four distinct lines of business that drive the industry.
The "Time & Material" (T&M) Engine
At its core, the traditional model is simple: Selling Effort. A client in New York needs a task done. Locally it costs $100/hour. An Indian provider does it for $25/hour.
Core Services ("Run the Business")
This is the industry's bread and butter—Application Maintenance & Support (AMS) and Infrastructure Management (IMS). It involves keeping mission-critical systems alive, 24/7/365. It is high-volume, process-heavy, and typically billed on a headcount basis.
Note: The BFSI (Banking, Financial Services, and Insurance) sector is the undisputed heavyweight champion here. Global banks spend billions annually just to maintain their legacy mainframes.
BFSI Example
Maintaining 30-year-old COBOL-based "Core Banking" ledgers. If these systems go down for even a minute, global money transfer stops. Indian engineers ensure 99.999% uptime.
Retail Example
Running the L1/L2 Helpdesk for a global retailer's Point-of-Sale (POS) systems. When a checkout scanner fails in a store in Ohio, a ticket is raised and resolved by a team in Bangalore.
Manufacturing Example
Managing the ERP databases (like SAP ECC) that track inventory across 50 factories. Ensuring the data flows correctly from the warehouse to the finance department.
Digital Transformation ("Change the Business")
This is the modern growth engine: Cloud Migration, Data Modernization, and App Development. Here, IT Services act as the Integrator in the value chain.
The Value Chain Reality
Clients buy the Infrastructure from Hyperscalers (AWS, Azure, Google Cloud) and the Software from ISVs (Salesforce, SAP, ServiceNow). They hire Indian IT firms to stitch it all together. The IT firm doesn't own the tech; they are the "glue" that makes it work.
Cloud Migration
Moving a bank's data from an on-premise basement server to AWS. This involves re-architecting applications to work in a modern, distributed cloud environment.
Customer Experience (CX)
Implementing Salesforce CRM for a global insurer so their sales agents have a "360-degree view" of the customer on an iPad, replacing clunky spreadsheets.
Data Engineering
Building a "Data Lake" on Azure for a healthcare giant, aggregating patient records from 100 different hospitals into a single, searchable format for analytics.
Platforms & Products ("The Pivot")
To break the linear link between revenue and headcount, companies are building their own IP (Intellectual Property). Instead of charging for hours, they charge a license fee or a "per-outcome" fee.
Banking Platforms
TCS BaNCS / Infosys Finacle: Core banking software products sold to banks worldwide. The bank pays a license fee to use the software, similar to how you pay for Microsoft Office.
Outcome-Based Services
Instead of billing for 50 people to process mortgage applications, the firm says: "We will process your mortgages for $500 per application." They use their own automation tools to do it efficiently and keep the margin.
Business Process Management (BPO/KPO)
Unlike traditional IT Services that build and maintain *technology*, BPOs and KPOs execute the *business processes* themselves. They supply the human capital to handle operational tasks—from answering customer calls to analyzing financial statements.
The Critical Distinction
"IT Services firms build the car (software). BPO firms drive it (operations)."
Customer Experience (CX)
Voice & Chat: Running 24/7 call centers for global airlines or telcos. Managing ticket resolution, refunds, and complaints.
Transaction Processing
Back Office: Processing millions of mortgage applications, insurance claims, or invoices. The focus is on speed and accuracy per transaction.
Knowledge Services (KPO)
High-Value Analysis: Financial research for investment banks, legal document review, or medical data annotation. Requires specialized domain knowledge.
The Pivot vs. The DNA
Critical Insight
Service providers are recognizing the imperative to pivot towards Outcome-Based Pricing and IP-led platforms. This transition is necessitated by the changing economic landscape, yet it presents a formidable structural challenge.
"It is not that this evolution is impossible, but rather that it fights against decades of organizational muscle memory. These firms have spent 30 years perfecting a machine optimized for labor arbitrage and linear execution. Transforming this well-oiled engine into one that fosters product culture and non-linear innovation is not just a strategy shift—it is a rewiring of their fundamental DNA."
2. The Productivity Shock
GenAI is not just a tool; it is a deflationary force striking the heart of the T&M model. Here is how it disrupts every service line we just defined.
The "Effort" Paradox
If an IT company's revenue is based on billing for hours (Effort), and GenAI reduces the effort required to write code or process a claim by 40%, then revenue theoretically drops by 40%. This is the central conflict. To maintain growth, companies must find 40% more work just to stand still.
The Mechanics of Deflation
1. Time Collapse
Traditionally, complexity equaled duration. A mainframe migration project was sticky because it was hard and took 18 months. GenAI automates the hardest parts—schema mapping, code refactoring, and test generation.
The Impact
Projects that anchored revenue for 4 quarters now finish in 2 quarters. The "Renewal Wall" hits faster, forcing sales teams to hunt for new deals twice as often just to replace churn.
2. Shorter Product Cycles
Code is becoming disposable. In the past, custom-built applications were assets maintained for a decade (generating 10 years of AMS revenue). With GenAI, rewriting an app from scratch is often cheaper than maintaining an old one.
The Impact
The long-tail revenue of "Maintenance" (often 60% of an IT firm's profit pool) is under threat. If code is cheap to generate, clients won't pay premium rates to maintain it.
Real World Evidence: The Efficiency Frontier
These examples are drawn from actual case studies and management commentary of leading IT service providers.
Core Risk Platform Modernization
What Happened
A leading IT firm rebuilt a mission-critical risk management platform in <6 months, a project that historically took 18 months.
Legacy Application Migration
What Happened
For a top-tier biotech firm, a service provider migrated a complex legacy application to a modern cloud stack 30% faster than manual benchmarks.
MVP Deployment
What Happened
Enabled a global entertainment giant to launch a new digital product feature (MVP) in record time.
Reporting System Modernization
What Happened
Transitioned complex financial reporting systems to a modern tech stack significantly faster than traditional methods.
Engagement Win
What Happened
Won a deal specifically because the solution offered faster release cycles along with productivity improvements.
The Threat from Above: ISVs & Hyperscalers
For decades, IT Services firms were the necessary "middlemen" bridging the gap between complex software and business needs. Now, the software makers are using AI to bridge that gap themselves, effectively squeezing out the integrator.
Hyperscalers
Companies like Amazon (AWS), Microsoft (Azure), and Google (GCP) that own the massive cloud infrastructure everyone rents. They provide the "digital ground" businesses stand on.
ISVs (Independent Software Vendors)
Companies like Salesforce, SAP, and Adobe that build the specific applications businesses run on. They create the "digital tools" employees use.
Exhibit A: Salesforce Agentforce
Replacing the "Implementation Project"
Before GenAI: A bank wants a customer service bot. They hire an IT firm (e.g., Infosys/Wipro) for a $2M project to scope, code, train, and integrate a custom chatbot. Timeline: 6 months.
With Agentforce: A business analyst at the bank simply types instructions into Salesforce: "Create an agent that can look up loan balances and process address changes." The platform builds and deploys the agent in minutes. No code. No IT project. No $2M fee for the service provider.
Exhibit B: Microsoft Fabric & Databricks
The Death of "Complexity Arbitrage"
Before GenAI: IT firms thrived on "Complexity Arbitrage." A client's data was messy—scattered across Snowflake, AWS, and SQL. IT firms charged millions to build ETL pipelines to stitch it together.
The Unified Future: Tools like Microsoft Fabric (OneLake) and Databricks (Lakehouse) are virtualizing data. They effectively say: "Stop moving data. Keep it here." By unifying data estates into a single SaaS layer, they eliminate the need for the complex, labor-intensive integration work that used to be a major revenue stream for system integrators.
How Incumbents are Responding
1. Strategic Pivot to Value Capture (Outcome-Based Models)
The industry is aggressively moving away from the commoditized "Time & Material" model towards Outcome-Based Pricing. This is a defensive moat against the deflationary pressure of AI. By contractually tying fees to business results (e.g., number of mortgage applications processed, percentage reduction in false positives) rather than hours worked, incumbents aim to capture the "productivity surplus" generated by AI. If an AI tool reduces delivery time by 60%, the provider retains the margin benefit rather than passing the cost savings entirely to the client. This shifts the negotiation from "cost per hour" to "value per outcome."
2. Expansion into Legacy Modernization (The "Brownfield" Opportunity)
GenAI significantly lowers the technical and economic barriers to modernizing complex legacy systems. Previously, rewriting millions of lines of "spaghetti code" (undocumented, archaic systems) was too risky and labor-intensive, often leading to project failure. AI-driven code analysis and refactoring tools now make these "brownfield" projects viable. This opens a vast, previously untapped market of legacy modernization work that was hitherto economically unfeasible, allowing incumbents to mine their existing client base for high-value transformation deals.
3. The Pre-requisite Boom: Data Infrastructure & Engineering
The immediate revenue opportunity lies not in AI models themselves, but in the infrastructure required to run them. "There is no AI strategy without a Data strategy." Enterprises are realizing their data estates are fragmented and unprepared for LLMs. This has triggered a massive wave of demand for "Data Engineering" services—cloud migration, data lake construction, data cleansing, and governance. Incumbents are positioning themselves as the essential architects of this foundation, driving a short-to-medium term revenue boom in Data & AI practices, often growing at 50%+ annually.
4. Structural Decoupling of Revenue and Labor
Perhaps the most profound shift is the conscious effort to break the "Linear Growth" constraint. For thirty years, growth was inextricably linked to hiring; to grow 20%, you hired 20% more people. Management teams are now explicitly targeting "Non-Linear Growth," where revenue scales faster than headcount. This involves a mix of higher billing rates for specialized AI talent, platform-based revenue, and the internal use of AI to drastically improve margin profiles. The metric of success is shifting from "Headcount Growth" to "Revenue Per Employee."
Strategic Signal: The Asset-Heavy Pivot
The Signal: TCS recently announced a massive ₹18,000 Crore (~$2.1 Billion USD) investment (with TPG) to build AI-ready Data Centers.
The Implication: This is a profound shift from the traditional "Asset-Light" services model to an "Asset-Heavy" infrastructure play. It signals that simply selling human services is no longer enough. To secure long-term relevance, the largest players feel the need to own the physical "ground" (compute & infrastructure) on which the AI economy runs, moving down the stack to capture value where it is most durable.
3. The Insourcing Wave: Global Capability Centers (GCCs)
A structural shift in how global enterprises consume Indian talent: Moving from "Rent" to "Own".
For decades, global companies (like Target, Tesco, or Goldman Sachs) outsourced their tech needs to third-party vendors (like Infosys or Wipro). This was the "Outsourcing" model.
Today, they are increasingly building their own captive centers in India—Global Capability Centers (GCCs). They are "insourcing" the talent, effectively cutting out the middleman to own the IP and culture directly.
Market Size (Revenue)
$64.6 BnGenerating massive export value, equivalent to ~25% of the total IT export market, but captured directly by global parents.
Hiring Velocity
4x FasterGCCs are expanding headcount by ~20% annually, compared to a sluggish 4-6% for traditional IT Services.
The Shift: From "Back Office" to "Headquarters 2.0"
Historically, GCCs were "Captive Centers" handling low-end support work (payroll, basic coding) to save costs. That era is over.
Today, they are Global Innovation Hubs. Fortune 500 companies are moving their core R&D, Product Management, and AI leadership roles to India. They are not here just for cheap labor; they are here for scale and talent quality. They are building the actual products (e.g., Uber's maps, Goldman's trading algorithms) from Bangalore and Hyderabad.
The Mid-Market Surge
It's not just the giants anymore. A massive wave of "Mid-Market" companies (Revenue $1B - $10B) are setting up smaller, agile GCCs (500-1000 people). They realized they can't afford the markup of an Infosys or TCS for their core IP. By setting up their own center, they get dedicated talent that understands their specific business domain, without the "rotation" risk of vendor staff.
Competitor & Client: A Complex Dynamic
IT firms actively help set up these centers. Using the Build-Operate-Transfer (BOT) model, a provider like Infosys will hire the team, set up the office, and run operations for 3 years before handing the keys over to the client. This turns a long-term threat into immediate revenue.
Running a GCC is hard. Many companies realize they lack the management bandwidth to handle Indian HR, compliance, and attrition. A significant percentage of GCCs fail to scale and eventually transfer operations back to IT service providers.
Providers are buying their way into this market. A prime example is Hexaware Technologies acquiring GCC-setup specialist SMC Squared for ~$120 Million. This allows them to capture the value chain upstream.
Read Deal DetailsThe Verdict: A Slow-Motion Collision
In the short and medium term, the impact on IT incumbents may appear muted. Revenue from BOT contracts and setup services effectively masks the underlying shift, keeping toplines stable.
However, the long-term structural reality is undeniable. GCCs are not just insourcing revenue; they are cornering the supply of premium talent. With deep pockets, higher pay, and the aspirational pull of global brands (e.g., "Working for Goldman Sachs" vs. "Working for a vendor"), they are steadily draining the intellectual capital that IT services firms rely on. Furthermore, GenAI usage will likely accelerate this trend, as critical AI workflows and productivity gains can now be insourced more effectively, reducing reliance on third-party operators.
The Timeline Paradox: Market vs. Reality
For Investors
Markets discount eternity in a second. Stock prices often move faster than reality. Investors need patience; structural shifts like GenAI adoption and business model pivots don't manifest in quarterly earnings immediately. The "death" of legacy IT is priced in long before the revenue actually disappears.
For Companies
Real-world transformation is heavy and slow. But for incumbents, the clock is ticking faster than ever. While the financial impact may lag, the risk of obsolescence accelerates daily. Resisting change is dangerous, even if the P&L looks fine for now.
4. The Incumbent Landscape
Tier 1 Giants: The Scale Defenders
Tier 1 Giants
- TCS
- Infosys
- Wipro
- HCLTech
- LTIMindtree
- Tech Mahindra
Financial Resilience vs. Growth Velocity
Aggregate Performance (Last 10 Quarters starting June '23)
A dual-axis combination chart titled 'FINANCIAL RESILIENCE VS. GROWTH VELOCITY' with the subtitle 'Aggregate Performance (Last 10 Quarters starting June '23)'.
The chart tracks financial performance over time using the following visual elements:
- X-Axis (Time): Spans 10 quarters from Jun 23 to Sep 25.
- Primary Y-Axis (Left): Measures 'Values in ₹ Cr' ranging from 0 to 2,00,000.
- Secondary Y-Axis (Right): Measures 'Margin (%)' ranging from 15 to 30.
The data trends are displayed as follows:
- Revenue (Light Grey Bars): Shows consistent, gradual growth. It starts at approximately 1,70,000 ₹ Cr in Jun 23 and steadily rises to nearly 1,90,000 ₹ Cr by Sep 25.
- EBITDA (Black Bars): Demonstrates stability, appearing to hover consistently around the 40,000 ₹ Cr mark across all quarters.
- Margin % (Black Line): Starts at roughly 22% in Jun 23, steps up to approximately 23% in Dec 23, and maintains a flat plateau until a slight dip in Jun 25, followed by a recovery in Sep 25.
The Moat: Scale & Complexity
Managing the IT backbone for a Fortune 500 company is not a task for a startup or an AI agent alone. The sheer inertia of legacy systems—Mainframes, complex SAP landscapes, and bespoke middleware—provides a massive buffer. These systems are the lifeblood of global enterprises; you don't "turn off" the engine that runs global payroll or high-frequency trading because of a new LLM release.
The "Trust" Barrier
BFSI Regulatory Liability
A global bank will not let a "black box" AI manage regulatory reporting. In a world of billion-dollar fines, the human-in-the-loop and legal liability provided by a Tier 1 firm is a non-negotiable insurance policy.
Life-Critical MedTech
When lives are at stake, validation is paramount. MedTech companies require deterministic outcomes that current generative models cannot guarantee without the rigorous auditing processes these incumbents have mastered.
The Pivot: Productization vs. DNA
Management teams have accepted the need to move towards outcome-based pricing. However, the attempt to productize their offerings remains a challenge. History suggests that IT services firms, optimized for human delivery, struggle to compete with world-class Independent Software Vendors (ISVs). The DNA of these firms is service delivery, not pure product engineering.
The Melting Cube
Legacy contracts act like a "leaking bucket." The base of low-complexity maintenance work—which historically provided high-margin cash flow—is being eroded by AI-driven productivity. Replacing this runoff requires winning new, high-value digital transformation deals at an accelerating pace.
The Amaltas Verdict
"While the financials demonstrate range-bound stability and suggest no immediate threat of disruption, it would be equally foolish to expect strong growth return. Whatever growth is captured through GenAI or productisation will likely be offset by the deflationary pressure in legacy contracts. With the market still valuing this cohort at ~25x PE, there appears to be a structural mispricing—the market is hoping for a return to growth that the underlying economics may no longer support."
Agile Specialists: The Growth Engines
Niche Operators
- Persistent Systems
- Coforge
- Happiest Minds
- Saksoft
- Hexaware
Accelerated Growth Trajectory
Aggregate Performance (Last 10 Quarters starting June '23)
A dual-axis combination chart titled 'ACCELERATED GROWTH TRAJECTORY' with the subtitle 'Aggregate Performance (Last 10 Quarters starting June '23)'.
The chart visualizes three key metrics over a timeline from Jun 23 to Sep 25:
- Primary Y-Axis (Left): Measures 'Values in ₹ Cr' from 0 to 9,000.
- Secondary Y-Axis (Right): Measures 'Margin (%)' from 10 to 25.
The data trends are displayed as follows:
- Revenue (Light Grey Bars): Displays a consistent upward trajectory. Revenue starts at approximately 5,100 ₹ Cr in Jun 23 and grows steadily quarter-over-quarter, reaching a peak of roughly 8,400 ₹ Cr in Sep 25.
- EBITDA (Black Bars): Shows parallel growth, starting just below 1,000 ₹ Cr in Jun 23 and rising to approximately 1,600 ₹ Cr by Sep 25.
- Margin % (Black Line): Exhibits volatility. It begins at 16% in Jun 23, spikes to ~18.5% in Dec 23, dips to a low of 15% in Jun 24, and recovers strongly to end at its highest point of nearly 19% in Sep 25.
Strategic Capability Acquisitions
Hexaware + SMC Squared
Entering the "Build-Operate-Transfer" market to counter the GCC threat.
Coforge + Cigniti
Acquiring deep quality engineering capabilities critical for validating AI models.
Happiest Minds + PureSoftware
Strengthening vertical depth in Banking & Healthcare product engineering.
The Amaltas Verdict
"Given the opportunities and risks, the aggregate currently seems to be overvalued. While the market is rewarding strong growth, it is perhaps not fully taking all execution risks into consideration. Although, if the tech transformation cycle really plays out, our view is that these companies will be the biggest gainers."
The biggest advantage that these companies possess is agility. Unlike their Tier 1 counterparts, they do not carry the enormous "legacy baggage"—revenue tied to maintaining 20-year-old mainframes or managing commodity helpdesks. Their business models were largely built during the Digital Transformation era, making them inherently more adaptable.
In the age of AI, digital transformation will likely accelerate, and these niche companies are positioning themselves to benefit immensely. They have actively sought to add ancillary business capabilities through strategic M&A (as seen in the adjacent panel) to offer specialized, high-value services that generalists struggle to match in speed.
Currently, this cohort is valued at an average of > ~40x PE, which is almost 60% higher than the Tier 1s. However, this premium has been earned on the back of extremely strong growth: a 26% CAGR over the last 2 years, against a mere 5% CAGR for the Tier 1 giants.
The Opportunity
Productivity Capture
These firms enjoy a medium-term window to capture AI productivity gains by moving to Outcome-Based Models. By delivering faster results with AI but charging for value, they can expand margins before clients demand the surplus.
The Key Risk
Vendor Consolidation
In a weak macroeconomic environment, enterprises consolidate vendors. This is a binary risk: specialists either win big by becoming the "Category King" or lose key accounts entirely to larger Tier 1 integrators offering bundled discounts.
Traditional Mid-Tier: The Squeezed Middle
Traditional Mid-Caps
- Birlasoft
- Zensar
- Mphasis
Stagnation & Volatility
Aggregate Performance (Last 10 Quarters starting June '23)
A dual-axis combination chart titled 'STEADY GROWTH, STRUCTURAL RISK' with the subtitle 'Aggregate Performance (Last 10 Quarters starting June '23)'.
The chart tracks financial performance from Jun 23 to Sep 25 using three key metrics:
- Primary Y-Axis (Left): Measures 'Values in ₹ Cr' ranging from 0 to 4,000.
- Secondary Y-Axis (Right): Measures 'Margin (%)' ranging from 10 to 25.
The data trends illustrate the specific scenario described in the title:
- Revenue (Light Grey Bars): Demonstrates a consistent, step-by-step increase. Revenue begins at approximately 2,500 ₹ Cr in Jun 23 and climbs steadily every quarter to reach a peak of roughly 3,700 ₹ Cr by Sep 25.
- EBITDA (Black Bars): Follows a similar steady upward trend, starting at around 500 ₹ Cr and rising to approximately 750 ₹ Cr by the end of the period.
- Margin % (Black Line): Reveals underlying instability despite the steady revenue growth. It starts flat at ~19% for the first four quarters, dips to ~18% in Jun 24, fluctuates up and down between 18% and 19% through Dec 24, and finally trends upward to reach roughly 20% in Sep 25.
The Strategic Trap
-
⚠
Vendor Consolidation Losing sticky legacy business to Tier 1 giants who offer better bundled pricing.
-
⚠
Innovation Lag Struggling to win digital transformation deals against Agile Specialists.
-
⚠
Margin Volatility High "Cost of Sales" as they hire expensive talent to pivot, destabilizing margins.
The Amaltas Verdict
"In our assessment, this cohort faces a particularly complex set of structural challenges. Confronted by both competitive consolidation and technological shifts, their traditional models are under pressure. While a successful pivot is possible, execution risks are elevated. Until clear evidence of a turnaround emerges, we believe the potential for long-term value erosion remains a key consideration for investors."
These are the companies that have been plagued with legacy models and have generally struggled for growth. They face structural problems due to a multitude of factors, with AI being a primary disruptor. Critically, they have not been able to truly transition away to digital transformations, which is where the real growth opportunities lie.
They face the "worst of both worlds"—squeezed between Tier 1s and Agile Specialists. They are not large enough to be immune to vendor consolidation deals, often losing legacy business to Tier 1 giants. Simultaneously, they lack the speed to compete with specialists for new digital mandates.
The "Leaking Bucket" Syndrome
"They are trying to fill a leaking bucket which may be disrupted by AI. To compensate, they are attempting to pivot their business models by increasingly investing in hiring expensive sales personnel or pursuing acquisitions. This has resulted in unstable margins."
The numbers reinforce this precarious position. Over the last two years, this cohort has delivered a topline growth of just ~7% CAGR—a figure indistinguishable from the Tier 1 giants, despite these companies being significantly smaller and theoretically capable of faster growth. This stagnation validates our thesis of structural headwinds. Yet, the market continues to assign a premium valuation of 25x - 30x PE, effectively pricing in a successful turnaround that has yet to materialize. Investors seem to be overlooking the underlying risks to the business model, placing faith in management's ability to pivot without tangible evidence of execution.
Process & Analytics: The Structural Risk
Process Leaders
- Latentview Analytics
- eClerx Services
- Firstsource
The "Factory Floor" of the Service Economy
To the uninitiated, the acronyms BPO (Business Process Outsourcing) and KPO (Knowledge Process Outsourcing) might sound like abstract corporate jargon, but they represent the engine room of the global service economy.
It is vital to distinguish them from the IT Services firms discussed earlier. If IT Services firms are the "Architects" who build the digital infrastructure (the apps, databases, and cloud servers), BPO and KPO firms are the "Operators" who actually run the business functions on top of that infrastructure.
What does this look like in practice?
They are the "human middleware." When you file an insurance claim, the person verifying your documents is likely a BPO employee. When a hedge fund needs to summarize thousands of earnings calls, a KPO team provides the analysis. They handle tasks that are too complex for old software to automate, but too repetitive or costly for high-wage onsite employees to perform.
BPO: "The Hands"
Business Process Outsourcing: High-volume, standardized execution. The goal is efficiency and cost arbitrage.
Real World Workflow: Firstsource
Input: A US bank sends 10,000 raw mortgage applications (PDFs) overnight.
Process: Agents in India manually verify income slips against tax returns, check for missing signatures, and key data into the bank's legacy system.
Output: "Verified" or "Flagged" status returned to the US loan officer.
Value Prop: "We provide specialized insights you lack the bandwidth to find."
KPO: "The Head"
Knowledge Process Outsourcing: Judgment-based analysis requiring domain expertise. The goal is insight and accuracy.
Real World Workflow: Latentview
Input: A global retailer provides terabytes of raw sales, inventory, and social media data.
Process: Data analysts use SQL/Python to clean the data, build statistical models, and identify buying patterns (e.g., "GenZ in Ohio prefers Red sneakers").
Output: A strategic dashboard for the CMO to allocate ad spend.
Value Prop: "We provide specialized insights you lack the bandwidth to find."
Steady Growth, Structural Risk
Aggregate Performance (Last 10 Quarters starting June '23)
The Core Thesis
"The Melting Ice Cube"
Unlike IT Services, which manages critical infrastructure and complex business functions resistant to short-term disruption, the BPO/KPO model faces Structural obsolescence. The threat here is immediate: they sell "Commoditized Human Intelligence"—a resource that AI is driving towards zero cost.
The Amaltas Verdict
"We observe that managements in this industry have been aided significantly by the strong post-COVID outsourcing wave, driven by acute labor shortages in the developed world. Additionally, they have actively moved to transform their business models through acquisitions, outcome-based pricing, and vendor consolidation."
"In the short term, these factors have provided a boost to both revenue and profitability growth. However, in the longer run, once large-scale AI adoption commences, we believe this sector will be the most vulnerable. At an average PE of 30+, markets appear to be assigning disproportionate weight to near-term growth while ignoring long-term structural risks, driven in part by positive management commentary."
1. Strategic Pivots & Capability Acquisition
Management teams in this sector have been candid in acknowledging the structural challenges facing the traditional Time & Material (T&M) model. In response, there is a concerted effort to pivot towards outcome-based models and productize service offerings. Furthermore, these companies are actively utilizing M&A strategies to acquire niche capabilities and accelerate their transition up the value chain.
2. The "DNA" Problem: Builders vs. Operators
The critical distinction lies in organizational DNA. IT Services firms are Engineers; their core skill is building systems. If tech changes from Mainframe to Cloud to AI, they retool the engineers.
KPOs are Operators. Their DNA is "Process." They hire graduates to follow a Standard Operating Procedure (SOP). When AI automates the SOP via natural language, the core skill—following instructions—becomes redundant. You cannot simply "fake" a tech culture pivot.
The "Product" Delusion
"Claims of pivoting to 'SaaS Platforms' are often illusory—typically just 'Service Wrappers' (basic UIs) that still rely on humans in the back. A true Tech company (like Salesforce or a specialized Fintech SaaS) will build a real AI product that solves the problem without any humans. The KPO cannot compete with that engineering firepower."
3. The Disintermediation Threat
Historically, banks outsourced because building custom automation bots was expensive. In the GenAI world, automation is easy. A large US bank can now use Microsoft Copilot or a custom LLM to automate "Level 1" support and document checking inside their own firewall for pennies. This secures their data and cuts the KPO middleman out of the loop entirely.
The Concluding Paradoxes: A Landscape of Contradictions
To conclude this study, we must confront the four central paradoxes that will define the next decade of Indian technology. These are not merely strategic hurdles; they are fundamental contradictions in the business models of incumbents that have yet to be resolved.
The Liability Paradox: Who "Owns" the Failure?
As the industry pivots toward Outcome-Based Models, it moves from selling "human effort" to selling "results." In the traditional model, if a project failed, the vendor lost a contract; in the outcome model, the vendor effectively becomes an operational insurer.
The Exposure
Consider a BPO processing 10,000 mortgage applications overnight. If a GenAI-driven agent misreads income slips due to a "hallucination," the liability isn't just a missed hour of billing—it is the financial and regulatory fallout of thousands of faulty loans.
The Balance Sheet Gap
Most Indian IT firms are "Asset-Light" and service-oriented. Their balance sheets are designed to manage payroll, not to absorb the massive liability shocks that come with owning critical business outcomes in sectors like BFSI or MedTech.
The Trust Barrier
Regulators often demand "human-in-the-loop" precisely because a "black box" AI cannot provide the legal accountability a Tier 1 firm historically offered.
The Efficiency Trap: Can Margins Survive Hypercompetition?
Firms hope to capture the "productivity surplus"—delivering work 40–60% faster with AI while charging for the value of the outcome. However, history suggests that in a competitive market, efficiency is a utility, not a moat.
Commoditization
If every "Agile Specialist" utilizes the same LLMs to drive a 30% efficiency gain, that gain becomes the new baseline. When the tool is democratized, the surplus is competed away.
Client Leverage
In a weak macro environment, enterprises consolidate vendors. Large clients are sophisticated; they will eventually demand that the AI-driven cost savings be passed back to them as discounts, rather than being retained as the provider's margin.
The Inertia Paradox: Betting on Corporate Sluggishness
A significant portion of the industry's profit—often 60%—comes from maintaining legacy "spaghetti code" and mainframes. This creates a bizarre incentive: earning stale cash flows from systems that are technically obsolete.
The "Melting Ice Cube"
Incumbents are essentially betting that corporate inertia and the "trust barrier" will outlast the deflationary force of AI. They are monetizing the friction of the old world while the new world accelerates.
The Disruption
GenAI makes rewriting these systems from scratch cheaper than maintaining them. As the "Renewal Wall" hits faster, firms relying on "Complexity Arbitrage" may find their "leaking buckets" emptying faster than they can win new deals.
The Capital Paradox: Dividends vs. Survival
For decades, the Tier 1 giants have been the "cash cows" of the Indian market, characterized by high dividend payouts. GenAI demands a radical shift from this "Asset-Light" history toward an "Asset-Heavy" future.
The Asset-Heavy Pivot
TCS’s ₹18,000 Crore investment in AI-ready data centers signals that simply selling human services is no longer enough; you must own the physical "ground" of compute to secure the value chain.
Investor Disconnect
This requires a fundamental shift in capital allocation. Are investors—who have grown accustomed to 80-100% payout ratios—prepared for a world where these firms must redirect billions into R&D and physical infrastructure just to stay relevant?
Final Verdict: From "Middlemen" to "Architects"
The Indian IT sector is no longer just "at a crossroads"; it is undergoing a cellular rewiring of its DNA. The legacy models of labor arbitrage and complexity arbitrage are being dissolved by the universal availability of cognitive compute.
The firms that survive will be those that transition from being "human middleware" to the indispensable architects of the AI economy. They must move from managing headcount to managing outcomes, from selling hours to selling intelligence, and from distributing cash to investing in survival.
For the rest, the risk is not a sudden crash, but a slow, quiet evaporation of value—the "Melting Ice Cube" in a warming world.