Daybreak - How India’s call centre industry is being rewritten by AI

Episode Date: October 21, 2025

At 2 a.m. in Bangalore, a call-centre agent is resolving flight refunds with a new kind of colleague — one that never sleeps. AI copilots are now embedded across India’s BPM sector, watching every click and keystroke to improve their own efficiency. For firms like Capgemini and Genpact, the real prize isn’t labour anymore — it’s workflow data. Because in the race toward “agentic AI,” whoever owns the data, wins. And India, for all its scaled up manpower, might be training the machines that will one day replace it.Tune in.Daybreak is produced from the newsroom of The Ken, India’s first subscriber-only business news platform. Subscribe for more exclusive, deeply-reported, and analytical business stories.

Hi, this is Rohan Dharma Kumar. If you've heard any of the Ken's podcasts, you've probably heard me, my interruptions, my analogies, and my contrarian takes on most topics. And you might rightly be wondering why am I interrupting this episode too? It's for a special announcement. For the last few months, I and Sita Raman Ganeshan, my colleague and the Ken's deputy editor, have been working on an ambitious new podcast. It's called Intermission.

We want to tell the secret sauce stories of India's greatest companies. Stories of how they were born, how they fought to survive, how they build their organizations and culture, how they manage to innovate and thrive over decades, and most importantly, how they're poised today. To do that, Sita and I have been reading books, poring over reports, going through financial statements, digging up archives, and talking to dozens of people. And if that wasn't enough, we also decided to throw in video into the mix. Yes, you heard that right. Intermission has also had to find its footing in the world of multi-camera shoots in professional studios, laborious editing, and extensive post-production.

Sita and I are still reeling from the intensity of our first studio recording. Intermission launches on March 23rd. To get alert, as soon as we release our first studio recording, episode, please follow intermission on Spotify and Apple Podcast or subscribe to the Ken's YouTube channel. You can find all of the links at the ken.com slash I am. With that, back to your episode. Meet Sham. He is a 25-year-old call center agent working in Bangalore. And at 2 a.m. in the morning, he isn't expected to think too hard. You see, his productivity, so to say, is measured by the minute handle time, which is basically just industry jargon for the average time

an agent spends wrapping up a customer call. Now, call centers have been using software that has been logging and calculating agents' minute handle time for years now. So, Shyam is no stranger to being tracked. But there's something new looking over his shoulder today. So yes, it's 2am and Sham is on his 19th query of that night. It is a flight ticket refund and it is resolved in six minutes flat. But not by him exactly. On his screen, a box is flashing prompts, taking notes, and drafting up a summary. Sham told us that he is always there. The he in question is an AI co-pilot who has been shadowing Sham's every call. And it's not just a productivity tool. Because well, every single tab Sham clicks, every claim he files isn't just resolving a query. It's all

also training data. And the AI that's watching his every move? Well, it's one that could replace him any day. But this story isn't just about sham. It's about thousands of service workers across India. It's about an industry that is simultaneously doing its job and rehearsing its last supper. And naturally, big global firms have taken notice.

For instance, back in July, KAP Gemini spent more than $3 billion, buying WNS Global, one of India's biggest BPM or business process management firms. But the price wasn't its call center staff. It was 30 years of workflow data, everything from insurance claims to refund approvals. And all of that data is now raw material for the next big thing in automation, Agentic AI. Agentic AI is not your average chatpot. It's a system that can run full operations from end to end by itself.

And the deal is a sign that India's next big export isn't trained manpower. Like sham. Instead, it's data. Welcome to Daybreak, a business podcast from the Ken. I'm your host Rachel Wilkes. And every day of the week, my co-host, Nika Sharma and I will bring you one new story that is worth understanding and worth your time. Today is Wednesday, the 22nd of October. For years, India's call centers were seen as the world's problem solvers on cheap labor.

In fact, NASCOM says India holds about 60% of the global BPM market. But with the advent of AI, the same outsource processes are also becoming proprietary products. If companies know how to repackage them, most still don't. A former member of WNS's AI team told my colleague the Ken reporter Mrun May that while WNS had all this process data, it wasn't really using it. And right now, everything in the tech world is moving so fast that if companies don't develop aggressively, they lose out. So that's where most of India's BPM industry stands today, caught between competing and consolidating. You see, global giants like Accenture, IBM and Oracle are spending billions on buying firms packed with workflow data and infusing AI into their operations. But India's BPM majors have been slower, with some exceptions.

like GenBacked for instance, which has made a few acquisitions in AI and analytics. But what's important to understand at this point is that the win here isn't just chat pods. Let's put it this way. If chat GPD is the eager intern reading from a script, then Agenic AI is the middle manager. It handles tasks, makes calls, files, reports and all without ever calling in sick. Dixionth D'Way, CEO of Zygmintzic and Agenic Customer Journey, platform told Mrunmi what he thinks the aim of Agenic AI is. It isn't to sound like chatwats, but to behave like a network of real employees.

It can go through all the motions of the workflow faster and with fewer mistakes. So for businesses, the pitch is obvious, and always on error-free workforce. Dowie says his clients have saved 200,000 hours of routine work and improved sales conversions by nearly 30%. And the more workflow data these systems ingest, the sharper they get, which is why everyone wants in and fast. So, BPMs which were once known for the headsets and graveyard shifts are now turning into analytics engines. An earthenSight, a Bangalore-based research firm, shared some data with the GEN.

It shows that analytics and AI now make up 40% of their business mix. This shift has created new players, like latent view, fractal, move, sigma, Tiger analytics and driven consolidation. In just five years, nine major deals have reshaped the industry. But the real race isn't just about who has the most call center agents anymore. It's about who owns the deepest pools of enterprise data. Like Gora Vasu of Unearth Insight told Mrunnay, whoever controls that data and can feed it into AI first wins.

But the thing is, the data rush began long before most Indian PPMs even realized. what they were sitting on, because they never thought of it that way. More on this in the next segment. Traditional BPM firms, as Gora Vasu tells us, never built technologies or systems to infer insights. They saw themselves as service providers, not product companies. And that complacency has left them vulnerable. You see, AI-native firms start with data, then layer technology on top.

Only now are BPMs realizing that they've had that data. all along. Now, the industry gap is already visible. Accenture saw it early. Back in 2020, it bought N3, an inside sales firm to strengthen Sinobs, its AI-powered operations platform. Today, analytics-led services

make up half of Accenture Ops' revenue. And analysts believe it's a slower players who will be acquired next. Gaurav Barab of Nelson Hall told us that basically tech firms aren't buying BPMs for just scale. they are also after the client base, and weak AI capabilities make BPM's easy targets. India's been here before. Its IT industry thrived on labor, but never made the leap to products and consulting.

Now, the BPM sector risks repeating that same story. As Bernstein India warned in a report, the industry is tearing down the barrel of AI-driven disruption. You see, the funding gap is stark. While Oracle and Accenture pour billions into AI, Indian AI startups raised just shy of $800 million in 2024. Their US peers, on the other hand, they raised almost $110 billion. But we have an exception in GenPact.

It was once a traditional BPM. But now, it's an AI-driven analytics firm. In 2024, nearly half its revenue came from its data tech AI segment, up from just 14% at WNS. Through acquisitions like Exponent L data, an AI-first data-driven solutions company, GenPact is shifting from solving complaints to predicting them. We spoke to Anirutha Ray, who leads agentic solutions at GenPact.

He told us that old BPM models could only take productivity so far. He said that true impact comes from combining process expertise with AI. Decades of running transactions for global clients have given GenPact deep operational knowledge, which gives it a unique vantage point, to embed AI directly into business processes and show measurable results. Most Indian BPMs, however, are still playing it safe. Stay tuned. The thing is, autonomous automation won't happen overnight.

Like Ziggmints survey said, they've already begun moving to AI co-pilots that help employees do their jobs better. For instance, an insurance agent can now generate premium estimates in seconds without waiting for an underwriter. The next step, he said, is for a writer. full agentic AI. But adoption is uneven. Now, adoption and sales

and marketing move faster because results show up in revenue. Dway said that it's easier to get companies to listen when they're shown top line growth. But convincing them about efficiency gains in claims or compliance takes much longer. The tech, on the other hand,

is improving fast though. The way comments that when Chachipiti first came out, it was bland. Now, it's better than an average sales rep. Not like one of the best ones yet, but getting there. What's surprising is that sometimes adoption starts

where you least expect it. For instance, one of Sigman's first clients was Nova IVF, a fertility clinic. The We had assumed fertility would be too sensitive. But it looks like women felt more comfortable talking to an AI because it didn't judge. So the way believes trust will grow slowly.

Just like how. people once learn to trust that ATMs wouldn't eat their cards. So, while big ticket sales will still need humans, the list of AI-managed tasks will keep on expanding. For workers like Sham, the change is already here. By dawn, after a six-hour shift, he had resolved 50 calls, claims filed, tickets refunded, with his AI co-pilot blinking by his side. As he sips a lemon ginger tea before heading home, he says that soon it will handle complaints by itself. The inevitability raises a larger dilemma.

India may hold the richest process data sets in the world. But without the know-how, capital, or urgency to exploit them, the back office of the world is drafting the instruction manual for its own replacement. Daybreak is produced from the newsroom of the Ken India's first subscriber-focused business news platform. What you're listening to is just a small sample of our subscriber-only offerings. A full subscription offers daily long-form feature stories, newsletters and a whole bunch of premium podcasts. To subscribe, head to the ken.com and click on the red subscribe button on the top of the Ken website. Today's episode was hosted and produced by my colleague Rachel Varghees and edited by Rajiv Sien.