India’s martech stack is exploding with tools, yet customer journeys feel more fragmented than ever. At the same time, AI, data privacy and ROI pressure are reshaping what it means to be a modern marketer. Sitting at the intersection of these shifts is CleverTap, a global customer engagement and retention platform built out of India.
In this conversation, Anand Jain, Co Founder and CEO of CleverTap, speaks with Brij Pahwa about why India is such a tough but exciting market, what “true personalisation” really looks like, how agentic AI will change marketing, and why CleverTap is focused on execution as it works towards an IPO.
We keep hearing that the customer journey is now completely fractured across tools and touchpoints. In this chaos, what is actually working in martech today, and where are the big gaps?
If you look at a market like India, the complexity starts with the consumer, even before you talk about martech.
We are fragmented by language, device, and network reality. Some people are on 2G, some on 5G, some on prepaid, others on Wi Fi. Every 100 kilometres the spoken language changes. English is not everyone’s first language. On top of that you have cultural diversity and a festival calendar that does not follow the Western calendar at all.
From a consumer’s point of view, they are just doing what is normal and natural for them. From a brand’s point of view, this makes behaviour hard to predict. Why do people buy at a particular time of day, or in a certain season, and ignore everything else?
On the brand side there are two big gaps. One, most companies buy martech in bits and pieces. They start with an email or SMS vendor, then add a pop up tool, then something else for segmentation. The data and the journey end up fragmented and chaotic. Two, there is a lack of expertise and a lack of an end to end view of customer data. Even at India or Asia scale, where apps can have hundreds of millions of users, many brands still do not have a unified view of the customer. That is the core problem.
Where does CleverTap come into this fragmented picture? How do you help brands stitch journeys, use data better and still protect user privacy?
The idea for CleverTap came from our own frustration in a large media organisation in India. We were the engineers asked to integrate all the tools and build dashboards that showed the customer journey. By the time we collected data, ran ETL and pushed it into a dashboard, the moment to intervene was gone.
So when we started CleverTap in 2013, the first principle was to unify the data layer. We could not find a database that did what we wanted at that scale, so we built our own data technology, which we call TesseractDB.
The idea is simple. A customer can push web data, app data, contact centre data, point of sale data, everything into one platform. That becomes a single source of truth.
Once you have that, you can do analytics, segmentation, prediction, recommendations, experimentation and personalisation from the same place. You can see where users are dropping off, what they are doing across channels, and you can intervene in real time rather than only visualising the journey.
This avoids the need for 25 different tools, reduces fragmentation, and lets the brand run its engagement in real time from one central brain.
There is much more ROI pressure on marketers now than in 2013. Which kinds of tools really deliver ROI, and how can brands in a fragmented market like India actually measure that ROI?
I would break ROI into two levels.
At a campaign level, marketers want better conversions, not just clicks. It is not enough that someone opened an email. You want them to transact, watch the movie, listen to the song, or order the food you recommended.
At a business level, you want to improve lifetime value. You want every customer to get what they came looking for, at the right time and in the right channel, so that they keep coming back.
If your stack is fragmented and you are just a point solution, it is very hard to show real ROI. An email vendor, for example, has no idea what happened on the website after the click. The web analytics tool does not know who sent the campaign. If WhatsApp, RCS, segmentation and analytics all sit in different products, you simply cannot put the full picture together.
The only sustainable way to drive ROI is to unify the core of your marketing stack on one platform. You might still use specialised vendors for last mile delivery, but the central nervous system for data, decisioning and journeys has to be unified and real time.
With something like TesseractDB under the hood, you can work at India scale in real time and move from broad targeting to a segment of one. When you combine that with experimentation and AI driven automation, you start seeing much better conversion and lifetime value.
Everyone talks about personalisation, yet we all still see irrelevant ads and spammy messages. From your experience, what does true personalisation look like, and what are brands still getting wrong?
In a perfect world, personalisation means matching five things correctly.
First, the right product for the right person. If you have a million users and ten thousand products, that match itself is a hard machine learning problem.
Second, the right copy. People buy the same product for different reasons. One person cares about the logo and status, another about the fabric, another about colour, another about getting the latest drop, another about price. The way you describe the product has to reflect that.
Third, the right creative. The visual should feel relatable. Age group, ethnicity, context, all of that matters.
Fourth, the right offer. For one person it might be a percentage discount, for another free shipping, for another loyalty points or wallet balance. You cannot give everyone the same thing.
Fifth, the right channel, timing and workflow. If someone has never shown intent, how many times will you reach out before you taper off, and at what times of day?
On top of that, real world context matters. Food preferences change if it is raining or if it is a festive or fasting period. External signals matter. Then there are internal issues like organisational silos, where different teams own email, banners and other touchpoints and do not coordinate, so the user sees different stories in different places.
The only way to solve this at scale is not manual targeting but automation. You need tools that mine your data, surface insights and then use both machine learning and generative AI. ML for matching users and products. GenAI for creatives and copy.
When you put this in an agentic framework and let the system optimise across all those variables, the spam problem starts to disappear. Messages feel like a friend who knows you well and recommends exactly what you were thinking about. That is true one to one personalisation.
FMCG and CPG brands often do not have rich first party data, and third party cookies are going away. Most of their sales data still sits with retailers. How can they build the kind of understanding and targeting you are describing?
In today’s world, if you do not have a direct channel to your customer, you are at a disadvantage.
There is a reason so many brands are going D2C. They want a direct relationship from brand to customer. When I am on your website or app, every click is a real time behavioural signal. Even if you are a shampoo brand, the filters I use and the variants I look at tell you something about my hair type and preference.
Brands that still behave like it is the 80s or 90s, and rely only on multi brand retail without owning any part of the digital journey, are missing that superpower.
The solution is to build digital touchpoints where customers can at least browse, if not always buy. Behavioural tracking on those properties allows you to understand needs, price sensitivity and preferences. Many retailers we work with do this very well and then extend it into email, WhatsApp and other channels to stay connected even when the customer is not on the site.
Owning that behavioural data is the foundation for everything else.
From your vantage point, working with so many brands, what unique insights have you gathered about Indian consumers?
Beyond the obvious language and device diversity, two things stand out.
One is trust and payment behaviour. In big cities, paying by card online feels natural. In many tier two or tier three locations, that level of trust is not universal. People may prefer cash on delivery, or want to see the product first. Brands have to respect that and design journeys accordingly.
Second is value and aspiration. India and much of the East are very value driven markets. People love getting a little more than they expected. The sabjiwala adding free dhania or the shopkeeper slipping in a small gift is part of our culture. Brands that do this thoughtfully build strong loyalty.
India is also an aspirational pyramid. Every layer wants to move up. That creates many niches and micro segments. Serving those aspirations well is a big opportunity.
If you look five years ahead, what is one bold prediction about how martech will change brand customer interactions?
I think the shift to privacy first and preference first marketing is happening even faster than five years. In the next six to eight months you will see much more of that.
Brands will have to respect not just data privacy, but also frequency and channel preferences. Some users will want one message a week or less. Some will prefer a particular channel. AI makes it possible to learn and adapt to this at scale.
If you combine large language models and machine learning in an agentic framework, marketing starts to feel less like “marketing” and more like a trusted friend recommending the right thing at the right time and price, in the right tone of voice.
Five years out, in a country like India with one and a half billion people, I see multiple micro niches, each potentially tens of millions strong. It will not be a winner take all market from a consumer brand perspective. The more we sandpaper the market into these niches, the better the experience for consumers.
I recently spoke with Rajesh from Netcore about twin agentic AI, with one agent on the customer side and a mirror on the brand side. At the same time, we have heard about CleverTap’s IPO plans. Where do those plans stand today, and how do you look at competition in this ecosystem?
On the IPO, we are executing on the plans, but we are still some time away.
The only way we earn the right to go public is by giving customers what they came looking for, building the best product, offering the best customer service and staying slightly ahead of the curve so that our customers get an edge over their competitors. When we are not out in public talking about it, we are heads down on execution. That is the focus.
On competition, we do not see Netcore as an enemy. I have known Rajesh for a long time and we are good friends. Strong ecosystem players like that actually make the whole space richer. The digital twin concept he spoke about is very much how we think about the future as well.
Related Articles
