Personalisation Beyond the Basics: How Kotak Life Engages Customers
In this interview, Prasad Pimple, Executive Vice President & Head of Digital Business Unit at Kotak Life, highlights his professional journey and strategies as a digital marketing leader at Kotak Life, focusing on the approach to digital transformation, personalisation, and multi-channel marketing in the insurance industry. He also explores how AI, automation, and CRM systems drive customer engagement and campaign optimisation.
Key Takeaways
- Digital transformation has significantly impacted insurance, with most processes now digital except physical medical underwriting.
- AI/ML is being leveraged for automated bidding in marketing campaigns, improving lead generation and reducing costs.
- Personalisation has evolved beyond basic demographics to include customer intent, content preferences, and preferred communication channels.
Multi-channel marketing strategies are crucial, with WhatsApp emerging as a prominent channel for customer engagement.
“Seamless integration of digital tools, CRM data, and analytics empowers marketers to deliver highly personalised and targeted experiences, optimising the customer journey and improving conversion rates.”
Can you share your professional journey and how your experience across various industries has shaped your approach to digital transformation at Kotak Life? What are your key responsibilities in leading the digital business unit?
Prasad: My career journey has revolved around digital marketing, beginning with Kotak Mahindra Bank post-MBA, where I first encountered marketing as a concept. At that time, customer communication relied heavily on physical direct mailers, used for engagement, retention, and cross-selling. However, with the telecom boom in India, digital communication evolved as customers began using personal devices, bringing SMS and email to the forefront. During my tenure at Kotak and ICICI, digital marketing centred around raising customer awareness, generating interest, and driving purchase intent through email and SMS.
When I transitioned to telecom with Vodafone and DoCoMo, SMS expanded to include Interactive Voice Response (IVR) and Outbound Dialers (OVD) for more interactive communication. This “voice plus digital” strategy became a standard in the industry, given the cost-effective nature of SMS and IVR for telecom companies. This personalised approach to customer conversations was foundational and informed my subsequent work in life insurance—a field historically low in engagement, where interactions were limited to policy renewals and occasional cross-sales. Today, however, the life insurance industry has transformed, leveraging digital communication not just for renewals but for meaningful customer engagement, enriching the overall experience with the brand.
Currently, at Kotak Life, I lead the digital business, which encompasses four verticals:
- Direct-to-Customer (D2C): Driving customer interest through digital campaigns, then converting it with contextual and trusted assistance.
- Online Partnerships: Collaborating with online brokers and web aggregators to make our solutions available to their customer bases.
- Digital Bancassurance: Integrating insurance products into net banking and mobile banking apps to facilitate DIY purchases for bank customers.
- Digital Alliances: Bundling small-ticket group insurance products within digital ecosystems, enabling seamless customer experiences for insurance purchases.
These verticals are supported by three enabling horizontals: e-Sales (contact centre operations), e-Ops (policy fulfillment), and Digital Technology (technical enablement for the digital business). My responsibility is to build and scale this digital business unit, positioning it as a primary channel for customer acquisition at Kotak Life.
Given that the consumer insurance journey—from evaluation to purchase to onboarding—is now primarily digital, how has this digital transformation impacted Kotak Life’s approach? Would you say the business is now almost entirely digital, or do offline components still play a significant role, particularly in certain areas?
Prasad: The life insurance industry has seen significant digital transformation, with most processes now conducted digitally. At Kotak Life, customer interactions, be it onboarding or servicing, are largely digital. Through our digital ecosystem, we’ve established DIY customer journeys for D2C businesses, enabling customers or intermediaries to complete the entire process digitally. Even field sales operations have transitioned to digital, from explaining product propositions to payment collection, unless large payments require physical processing.
While financial underwriting is predominantly digital, certain aspects of medical underwriting, such as physical exams, remain offline for term insurance products due to the need for thorough assessments. Although some enablers, like home visits and digital report transfers, are in place, physical sample collection and analysis are still necessary. As advancements in technology, like facial scans for vital checks, continue to evolve, these processes may also become fully digital in the coming years.
On the customer service side, nearly all interactions have shifted to digital channels, with WhatsApp playing a central role in addressing customer queries. Digital transformation in life insurance has progressed significantly, yet there remains scope for further enhancements to fully optimise the industry’s digital journey.
Read More
How have advancements in technology like automation, AI/ML, and analytics influenced the way you manage campaigns across various channels? Can you provide examples of how these technologies have enhanced personalisation, scalability, and collaboration within your campaigns?
Prasad: Technology has become an essential component of modern marketing, enabling precise targeting of audiences with the right content at the right time. Digital marketing and analytics integration now allow for comprehensive tracking of customer interactions. Unlike traditional media, where tracking the impact of a TV commercial was challenging, digital channels provide visibility into every step of the customer journey—from seeing an ad to visiting a website, interacting with a page, filling out a lead form, and ultimately deciding to purchase. This data empowers marketers to personalise and tailor communication based on the customer’s unique journey.
Analytics plays a critical role in what is now known as full-funnel marketing. With full-funnel strategies, marketers can engage customers at every stage, from building initial awareness to encouraging consideration and ultimately driving conversions through retargeting. Advanced marketing technology and CRM systems have further enhanced this approach, integrating data from digital interactions and customer service touchpoints into one centralised system.
For example, analytics may reveal that a particular type of prospect converts better through contact centre interactions. By capturing this information in the CRM and feeding it back into marketing tools, marketers can create lookalike audiences to attract similar, high-converting prospects. This seamless integration of digital tools, CRM data, and analytics empowers marketers to deliver highly personalised and targeted experiences, optimising the customer journey and improving conversion rates.
AI and machine learning are increasingly integral to marketing, with most platforms now leveraging these technologies. For instance, in Google’s ad bidding system, multiple optimisation methods are available, such as click optimisation, lead-focused bidding, and Target CPA (tCPA). Traditionally, a human would analyse campaign data periodically, gain insights, and adjust bidding strategies accordingly. However, this approach limits responsiveness due to the time needed for manual analysis and decision-making.
To enhance efficiency, we experimented with an AI-driven automated bidding project for our search campaigns, enabling real-time decision-making. The machine learning model continuously analyses campaign data, identifying customer segments with higher conversion rates and reallocating bids accordingly. This real-time optimisation led to a notable increase in converting leads while significantly lowering the cost per lead.
While AI and machine learning are already transforming our marketing strategies by improving customer targeting and reducing costs, this is just the beginning. As new tools and ecosystems emerge, we expect even broader applications of AI in marketing automation and audience targeting, further enhancing campaign performance and precision.
What are the major challenges you’ve encountered while implementing marketing automation?
Prasad: Starting an automation or machine learning project brings a high potential for challenges, as the process often requires significant time for the technology to learn and adjust before yielding optimal results. This initial phase demands patience, as machines need time to interpret data accurately and make informed decisions.
In any new automation or marketing campaign, beginning on a small scale is crucial. Rapid scaling can lead to unforeseen problems, as we have learned from past experiences where quick expansion backfired. A carefully controlled approach, with continuous manual oversight, allows for real-time monitoring, helping to catch issues early. Waiting days or weeks to identify problems can lead to unnecessary setbacks and can risk a promising project failing due to execution flaws rather than conceptual ones.
There are three primary challenges in these projects:
- Setting Realistic Expectations: Often, expectations are set too high from the outset, leading to disappointment if results fall short. Clearly defining the project’s objective and expected outcomes helps set a grounded vision, making it easier to assess success accurately.
- Preventing Execution Errors: Close monitoring during the initial rollout phase is essential. Without oversight, execution inefficiencies can creep in, causing the experiment to fail not because of a conceptual problem but due to missteps in implementation.
- Ensuring Scalability: Even if a project works well in its initial phase, scaling up can present its own challenges. We’ve observed that some solutions, while effective on a smaller scale, may not transition well to a broader application due to design limitations.
To navigate these challenges, our approach focuses on three foundational elements: first, setting clear, realistic objectives to align everyone’s understanding of the project’s aims; second, ensuring flawless execution, even if it requires additional time, as getting it right matters more than being first; and finally, evaluating scalability from the beginning. By confirming that a successful experiment can be expanded smoothly, we increase our chances of long-term success. This methodical, step-by-step approach provides a solid foundation for effective, sustainable automation and machine-learning initiatives.
Personalisation is a critical aspect of customer engagement. How do you scale personalised marketing efforts across a vast customer base while ensuring relevance and impact?
Prasad: In marketing, personalisation and customisation are now essential to engaging customers effectively. Personalisation has advanced well beyond merely addressing a customer by name. Today, it’s about understanding the customer’s intent and crafting experiences that align with their unique journey and needs.
For instance, if a customer begins the journey to purchase term insurance on our platform but drops off midway, we can identify their interest in this product. When they revisit our site, the landing page and banners can get tailored to guide them seamlessly back into the process, highlighting their selected product and offering support if needed.
For those who don’t return, we’ve established a personalised “chase” marketing strategy. Initially, a message is sent offering assistance, allowing them to select a time for a follow-up. If there’s no response, we will send a second message outlining the benefits of the selected product and provide options for further inquiries via callback. This continues with reminders that emphasise the product’s unique features, discounts, and ease of purchase. If the customer remains unresponsive, we shift focus, suggesting other solutions like insurance for child education, retirement, or other goals. This approach ensures that each customer receives a personalised experience, whether they purchase or explore alternative solutions.
Beyond personalisation, customisation in content is becoming increasingly vital. Previously, marketing teams based content on demographics, assuming people of similar age, location, or gender would respond similarly. However, content consumption preferences vary significantly, even among individuals with shared demographics. For example, while one person may prefer emotionally driven content, another may gravitate toward data and statistics. Similarly, content format matters: some may prefer written articles, finding them more authoritative, while others favour quick, explanatory videos.
The future of marketing lies in understanding these nuances in content preferences—both type and format—and delivering tailored content that aligns with each customer’s unique consumption style. Though marketers are still gathering the data needed for this level of customisation, the goal is clear: to refine communication platforms so that content reaches customers in their preferred format, enhancing engagement and relevance.
With audience preferences varying not only by age or demographics but also by content type—emotional vs. data-driven—and format, such as text or video, how do you manage this in a multi-channel landscape? Given that channels like WhatsApp are currently prominent, but others like Instagram may become more significant over time, how do you approach audience segmentation, content selection, and ROI measurement across these evolving platforms?
Prasad: To navigate the complexities of multi-channel marketing, we segment our approach based on customer engagement stages and preferences. Platforms like Meta and Google are key to prospecting new customers, as many are actively searching for solutions there. The format is often experimental; however, video works well for building awareness, while static banners and emails drive consideration, clicks and intent.
When engaging customers who have already shown interest, email remains effective due to its ability to provide comprehensive information non-intrusively. SMS has seen declining relevance, especially as it’s now primarily used for transactional alerts like OTPs. WhatsApp, however, has gained prominence as both an interactive platform for queries and a robust channel for communication. Its versatility allows for multiple content formats, and though it’s more costly than SMS, it provides a closer, more responsive customer interaction.
For instance, when running term insurance campaigns on Meta, rather than redirecting users directly to a web page, we now often link them to a WhatsApp bot. This bot answers initial questions, provides quotes based on customer details (like age and gender), and gives a tailored experience. This approach improves lead quality since customers engaging through WhatsApp and receiving personalised responses are generally more interested than those who simply click on a web page.
In summary, integrating multi-channel marketing with tailored content formats across platforms like Meta, Google, email, and WhatsApp allows us to enhance customer interaction and increase lead quality by matching each stage of engagement with the most appropriate channel and communication style.
Lead attribution plays a key role in optimising resources and maximising ROI. How do you approach lead attribution in your campaigns? Could you walk us through the attribution models and stages you use to identify which channels or campaigns are most effective?
Prasad: There are several approaches to attributing a lead conversion, with many brands opting to assign equal weight to each touchpoint in a customer’s journey. However, we’ve found that focusing on the first source provides the most meaningful attribution for a conversion. Analysing customer paths reveals that initial platforms like Meta, Google, or native platforms play a significant role in the customer journey.
While customers may receive nudges along the way (such as an email reminder after initially engaging through Meta), attributing conversion weight to each nudge often complicates analysis without adding clarity. For us, giving primary attribution to the initial source aligns better with our overall marketing goals.
However, we don’t follow a rigid model for business attribution. Instead, we combine both first-touch and last-touch attribution, acknowledging both the platform where a customer first engages and where the final conversion occurs, ensuring a balanced view of the customer journey.
To wrap up, could you share your key expectations for this report? As a functional leader, what insights or focus areas would you find most valuable to help plan your campaigns more effectively?
Prasad: My main expectations from this report are twofold. First, I’m keen to understand the experiments and strategies currently used across different industries, especially those that have proven successful over time. Learning how different sectors, such as e-commerce or FMCG, approach customer engagement for varying goals—whether building brand awareness or driving action—would provide valuable insights into what works and what doesn’t for each marketing objective.
Second, insights on customer behaviour changes, particularly those observed by other industries, would be highly beneficial. For instance, in digital marketing, there’s often a crossover between online and offline influence. Understanding how other businesses perceive and measure the impact of digital marketing on both channels and whether they view it through an omnichannel lens would be invaluable for enhancing our own approach.
