Begin typing your search...

Data consumption for product innovation

Leveraging data consumption is key to driving product innovation, enabling companies to understand user behavior, predict trends, and create more personalized, efficient, and successful products.

Data consumption for product innovation

Data consumption for product innovation
X

31 July 2025 11:58 AM IST

One of the challenges the insurance industry faces is that “it remains a traditional industry, despite the introduction of numerous technologies,” says Ayan De, Chief Product Officer, Zopper in an exclusive interaction with Bizz Buzz.

What are the key challenges that organisations face when it comes to building scalable customer-centric and data-driven products?

One of the challenges the insurance industry faces is that it remains a traditional industry, despite the introduction of numerous technologies. It has not yet fully integrated these technologies into its general operations or the way the industry moves forward. Still, the basic structure is a very traditional one, and that is not wrong with it, because we deal with a lot of risk-taking, and that has been the cornerstone of all strategic decisions in the past as well. Now, with that situation, there have been different phases in which technology has evolved in this industry. Additionally, it has also been due to the typical nature of the products that the industry sells. It has also been, you know, very much human-centric as well. So it was never a case where technology completely took over the human element. As a result, the growth of technology in this sector has become somewhat siloed over the last three or four decades. Now, the challenge is that the oldest stack in this entire framework is the core policy admin system, which is essentially the core banking system, serving as the nucleus of the whole thing. That was possibly the first which got developed in the industry. Over the years, a significant amount of business intelligence has been integrated into that. Now, the challenge is that while you can extract data and bring it forth for consumption and ingestion by AI, AI engines, how do you read through the business logic and business algorithms that have been built over the decades? None of the AI, the ones we typically deal with, are still powerful enough or intuitive enough to read beyond the data points and harvest the business logic and intelligence that has been built in. That's again one of the bigger challenges that the industry faces today.

What is the importance of data security, data accessibility and data compliance – and the role enterprises are playing in addressing these three pillars?

Digitisation is transforming every aspect of the insurance industry, from policy underwriting to claims processing and customer experience. Innovations in AI, automation and data analytics are redefining the underwriting process flow. All this also calls for scrutiny, as data plays a pivotal role, and even a slight compromise in data can lead to significant defaults, resulting in a poor customer experience. Data security, data accessibility, and data compliance are crucial for building trust and ensuring the responsible handling of sensitive information. Companies must pay detailed attention towards encryption, monitoring and accessing data when it comes to customer service. Governments and industries are exploring ways to regulate AI and its applications; however, AI is evolving rapidly. Evolving data privacy regulations are shaping the way businesses approach compliance this year. This is where the power of cyber insurance comes in. Historically, cyber insurance is a relatively new product. Twenty years ago, no one was talking about it. Ten years ago, the concept took shape. Just five years ago, the market truly emerged.

Why do you think that there is a need for data-driven decision-making when designing products that meet customer needs and demands?

The insurance sector is aligning with the rest of the financial world in its approach to data strategy and the rise of AI, primarily driven by concerns over data privacy, security, and compliance. Additionally, this is just one side of the story; the proliferation of AI technology, per se, is also required at the other end of the spectrum. Requiring a lot of structured data to be available for the AI engines to consume.

Larger enterprises tend to make decisions based on data rather than intuition. What has changed now is, first of all, the amount of data that can be harnessed. Number two is the real-time availability of a lot of this data. And number three is, with so much data, I mean, the insurance industry sits on a mountain of data, with AI now getting into that, you know, bringing all the dots together and creating insights out of this data mountain that has changed the world.

Cloud computing has been a game-changer for the insurance industry for several years. Given the larger customer base from both the partner and end-user perspectives, there is always a strong sense of reliability regarding data storage in the server. We utilise embedded APIs to enable insurers to distribute insurance products through a vast network of ecosystem partners, thereby expanding their reach and reducing distribution costs. We have enabled auto-scale mode for all our programs, ensuring they are written in a way that allows them to scale automatically. This means that the moment a spike occurs and a requirement for different resource types from the server arises, they can adjust accordingly. It can be a memory. It can be a computer. Therefore, there is a clear correlation between the cost of technology today and the adoption of cloud services. And that has also enabled us to expand our business. Typically, a cloud-first approach enables insurance providers to leverage SaaS (Software as a Service) and PaaS (Platform as a Service) solutions with minimal investment and downtime. A complete technology transition to the cloud or PaaS/Saas integration offers similar benefits to insurers. These include accelerated response time to emerging market needs, scalability to expand or shrink on demand, reliability with uninterrupted data and technology availability, and comprehensive security.

We can now have businesses run marketing campaigns day in and day out, whenever, in the middle of the week, at the end of the month, or whenever our partners or clients demand, and the infrastructure is always available.

How are organisations driving a balance between AI-driven personalisation when it comes to customer-driven product development, and also ensuring data privacy and regulatory compliance?

When we talk about AI, we typically tend to think more about the LLMs in the public domain, such as ChatGPT, and the LLMs in the world, as well as the deep learning models. But they are more for public consumption, for the public domain, in a more controlled environment and a highly regulated environment of a financial company. You cannot afford to. Practically use any of these public models. So, the moment you develop those public models and stay within the walls of your own specified language models, which are clearly defined and designed for the industry, you can achieve this. And you have your goals, specifically product personalisation and personalised services, of that nature. We are seeing them evolve and develop hand in hand. Because one is there, the other is happening. It would not have occurred in isolation today. You cannot provide a personalised CX experience altogether without the power of AI running in the background. So, this is not a balance. I would say this is more complementary today.

What are the key factors that enterprises should consider when scaling AI and cloud adoption for long-term success in the insurance sector?

AI models, anyway, are built to scale. The challenge for the insurance sector is that I see many of our old tech stacks; they need to move ahead quickly and upgrade where they can leverage the potential of AI and cloud. Cloud and AI per se are scalable technologies. It is now for the insurance industry to decide how quickly they want to move away from their core systems, adopt newer technologies and platforms, and then they will be in a better position to assess the scalability of these technologies. I'm flipping the question around, actually, because that's the truth today.

EoM.

data consumption product innovation data-driven development user behavior analysis predictive analytics product design insights big data customer data innovation strategy product improvement data analytics in product development 
Next Story
Share it