From 65 Lakh Hinglish Queries to 29 Million Business Enquiries: Nikhil Prabhakar, CIO, IndiaMART, shares his digital playbook

India’s B2B marketplace landscape is undergoing a seismic shift, driven by the country’s rapid digital adoption and the increasing need for SMEs to transact seamlessly online. With over 8.4 million seller storefronts, 119 million products across 56 industries, and 29 million unique business enquiries annually, IndiaMART has become the digital lifeline for millions of businesses navigating this transformation.

But scale alone isn’t enough. In an ecosystem where 65 lakh queries every month come in Hinglish, misspellings, or vernacular languages, the challenge lies in making discovery seamless, engagement frictionless, and trust uncompromising. This is where technology leadership becomes critical.

Nikhil Prabhakar, CIO of IndiaMART InterMESH Limited, shares how IndiaMART is future-proofing its platform through AI-led discovery, conversational commerce, catalog intelligence, automation, and security-first practices—while preparing for the next wave of B2B tech that could redefine how India’s businesses connect, collaborate, and grow.

Some edited excerpts:

What are the cornerstone digital transformation initiatives at IndiaMART to ensure the platform remains future-proof for its 8.4 million seller storefronts and growing buyer base, particularly in the context of India’s digital-first economy?
At IndiaMART, our digital transformation strategy is about making B2B commerce borderless, language-less, and timeless for our 8.4 million sellers in India’s digital-first economy. We translate this vision into action through 5 cornerstone initiatives:

AI-led Search & Matchmaking: Our platform handles ~65 lakh Hinglish and misspelled queries monthly. Currently, the platform is enabled with 9-language voice search, AI relevance models that further improve the CTR & lead quality.

Catalog Intelligence: Our digital-first strategy enables the sellers to use multiple AI tools to create an intelligent and visually delightful catalog using automated background removal, cropping, attribute tagging, and fraud/spam takedowns at scale.

Conversational AI: Today, more than 60% messages are sent using AI “Suggested Replies”, helping the sellers to effectively communicate with customers. Our latest product IM Insta helps sellers directly deliver their replies to buyers on WhatsApp, ensuring ~3× higher buyer responsiveness.

Scalable, Open Infrastructure: Our team’s constant work on dual/multi-cloud, microservices, Kubernetes/Docker, Golang/PostgreSQL stack helps in faster deployment cycles and scale to millions of concurrent users.
Security & Compliance: Given the scale at which we operate, we also ensure that we are fully compliant with current regulations (ISO 27001/27701/22301/31000/12207 compliance), have the right certifications and security infrastructure in place, including RBAC, MFA, encrypted PII, WAF, and continuous VAPT.

Collectively, we ensure IndiaMART not only scales seamlessly with India’s digital-first economy, but also delivers faster discovery, higher engagement, and stronger trust for every buyer–seller interaction.

How is IndiaMART leveraging AI and machine learning to enhance operational efficiency, personalize user experiences, and drive the 29 million unique business enquiries, and what specific AI-driven tools have proven most impactful?
For a platform as huge as ours, AI and ML are no longer pilots. They are now embedded in our core workflows to make discovery faster, engagement richer, and operations sharper. Our goal is to use AI where it moves the needle for buyers, sellers, and our teams. Some of the core areas where we are making a difference with AI include:

Discovery: Our AI models handle ~65 lakh misspelled/Hinglish queries every month, across 9 Indian languages, reducing null results and improving relevance.

Behavior-led Recommendations & Matchmaking: Our ML algorithms use buyer intent signals to ensure sellers receive leads of relevant buyers. From suggesting a behaviour location-led buyer to understanding whether the location, product, or the order value is of interest to the seller, the algorithm personalizes lead recommendations as per the seller & vice-versa, making it a two-sided match.

Trust & Risk Management: Our AI models flag suspicious listings, block spam leads, and catch policy violations, ensuring fraud and spam detection at early stages.

Specification & Attribute Extraction: Our NLP models can read unstructured data to auto-fill product attributes, which ultimately helps in reducing seller onboarding effort and ensuring the seller catalog has ample information

Today, IndiaMART handles more than 29M+ unique business enquiries annually, with higher match relevance through AI. Our first-response times have been cut by over half for AI-assisted sellers, and null search results have been reduced significantly for vernacular/Hinglish queries. For us, AI is not an add-on, but an essential part of our core functionality and operations, and it has been for a couple of years now.

With IndiaMART’s vast MSME ecosystem and increasing digital transactions, what advanced cybersecurity protocols and data privacy measures are in place to protect user data and maintain trust?
Protecting user data is as important as maintaining trust across our massive MSME ecosystem. Security is embedded at every layer, from infrastructure to user interactions, though explicit certifications (like ISO) haven’t been publicly confirmed. This is part of our security-first philosophy.

Infrastructure and Operational Security: We operate on secure, cloud-native architecture, leveraging industry-standard best practices like WAFs, multi-layer firewalls, intrusion detection, and strong encryption protocols. The platform emphasizes high availability and resilience to ensure that buyers and sellers can transact securely at scale.

Identity & Access Management: We have robust access control measures in place, including principles akin to Role-Based Access Control (RBAC), Multi-factor authentication, and access segregation for critical systems (common patterns but not formally confirmed by IndiaMART).

Data Privacy & Minimal Exposure: Data handling is structured such that only necessary data is retained, in compliance with standard privacy norms and user expectations. Especially, the personal and payment-related information is handled under a minimized data exposure policy.

Risk Monitoring & Incident Readiness: We also maintain continuous monitoring and alerts for anomalous activities, as early detection systems are foundational for large-scale marketplaces like us.

How has automation been implemented to support scalable backend operations and tech-enabled lead generation, and what tangible benefits are sellers and buyers experiencing as a result?

IndiaMART handles more than 29M+ unique business enquiries annually. We continue to experiment and make changes in the platform to enhance the user experience and buyer-seller journeys on the platform. Some of our initiatives to enhance the platform experience for users include:

From a buyer’s perspective:
Lead quality: In the process of submitting a requirement, buyers are nudged in their journey to provide specifications, quantity metrics to improve better matchmaking, thus improving seller experience. Personalised Recommendations based on search behaviour and past interactions. Searches in Hinglish, misspellings, across nine Indian languages, including voice-based queries

From a seller’s perspective:
AI-Prioritized Leads: Instead of just a list, the dashboard highlights the recommended lead that our AI has scored as high-intent based on the past behavior, allowing the seller to focus their efforts.

One-Click Actions: Sellers can send payment reminders, share their catalogue, or respond to an inquiry with pre-set templates using a single tap, all automated.

Catalog management: IndiaMART hosts more than 119 million products across 56 industries; therefore high-quality catalog is imperative in order to ensure that buyers get all the information at the earliest. We have implemented an innovative ensemble model that combines a deep learning model with two index-based models. This approach automates product classification with minimal human intervention, leading to significant gains in efficiency and accuracy. As a result, we’ve been able to handle the vast scale of our catalog more effectively.

Product-led enhancements: We are also working on creating standard products for better specifications to minimise the manual efforts of filling up specifications. Under this, the seller can select the product they are adding to their catalog, and it will automatically fill up the standard specifications of the product.

Chatbot support: To further enhance the user journey, we are also introducing a chatbot and voice bot for sellers and buyers. With this, a buyer can simply state his/her requirement, and the bot will automatically create a buying requirement for the buyer. Additionally, the voice bot will also support users in resolving common platform queries and connect them to the customer agent, wherever needed.

How is IndiaMART utilizing data analytics to gain insights into buyer and seller behavior, and how are these insights shaping product offerings and platform strategies?
We are not a data-led, but data-enabled organisation. We believe that data analytics is not just about reporting, but turning millions of daily interactions into actionable intelligence that shapes our product roadmap and platform strategy. Our data analytics can be divided into two cohorts: buyer and seller.

Understanding Buyer Behavior

Search & Enquiry Analytics: These track what buyers are searching (including Hinglish & misspellings) and where searches lead to no results, enabling us to expand product categories and improve search relevance.
Journey Analysis: It identifies drop-off points in the buyer journey and introduces nudges such as prompting buyers to provide specifications or quantities. This can improve lead quality for sellers.

Understanding Seller Behavior
Responsiveness Metrics: It monitors how quickly and effectively sellers respond to leads, and then uses the data to recommend features like Suggested Replies or IM Insta to boost engagement.

Product Quality Scores: Evaluate listing completeness, image quality, and product attribute accuracy; nudge sellers to improve their products & catalog for higher visibility.

Conversion Tracking: It analyses which seller behaviours lead to successful closures and feeds this back into seller dashboards & helping the sellers monitor and change their behaviour.

How do we use them: We turn these insights into actions during our development processes.
Product Development: These insights guide new features across all products.
Personalization: Better analytics for sellers to help them grow their business and improve their operations for better efficiency

Over the years, these actions have helped us create a positive impact on the user journey, including higher lead quality, conversion rates, improved buyer satisfaction through faster, more relevant matches, and increased seller success by aligning tools and features with proven high-performance behaviours.

With ₹2,762 Cr in cash reserves, what does IndiaMART’s technology roadmap entail for the next 2-3 years, particularly in areas like AI, data analytics, or emerging technologies such as blockchain or IoT?

We remain committed to bringing innovations in the B2B space and creating an ecosystem that supports seamless conversation between business buyers and sellers. We also believe that we have barely scratched the surface when it comes to AI/ML integration within our workflows, and there is a huge potential to explore that area. Some of the focus areas include:

Advanced Seller Analytics Suite: We will invest in building a premium analytics suite. This will provide our sellers with deep insights into their competitive landscape and pricing intelligence. It’s about giving every small business owner the kind of data-driven insights that were previously only available to large corporations.

AI as the Central Nervous System: AI will not be limited to today’s search, lead generation, or suggested replies; it will evolve into the core intelligence layer of IndiaMART. With new use cases spanning across customer experience, seller onboarding, and productivity gains.

With one of the largest B2B data sets in the country, our roadmap includes transforming this data into a direct asset for our users and partners and building a Data as a Service (DaaS) Ecosystem.