Inside IndiaMART’s evolving approach to smarter commerce

India's digital commerce ecosystem continues to grow at a rapid pace, and companies such as IndiaMART that serve a broad user base (with sellers and buyers) are looking to artificial intelligence (AI) to address scale, serve users better, and improve security.

Platforms like IndiaMART have millions of queries recorded each month - often even typed in different languages, spellings, or sometimes slang - presenting a significant opportunity for platforms to be more efficient while still returning accurate results.

In this spirit, Dataquest spoke with Nikhil Prabhakar, CIO of IndiaMART InterMesh Ltd, about the new ways the company is employing AI and generative technologies in its platform, regarding search, cataloging, fraud detection, and its internal operations while needing to achieve the balance between innovation with responsible governance and the protection of data. Excerpts: 

Can you give one or two easy examples, how does IndiaMART using AI, or Generative AI to improve things for users?

We have been using AI for a much time, well before it became a buzzword. IndiaMART is an evolving, maturing company that is constantly trying new projects and cases; to make users' experiences better when using the site. A couple of areas where AI technology has been integrated greatly include: a more efficient completion of product cataloging, making the esthetics of listings more appealing, and creating more engagement. 

• Reading customer behaviour: Generative AI analyzes the buyer & seller behavior and preferences on the platform, and suggests tailored recommendations for products or suppliers, making the platform more personalized, intuitive, and responsive to the specific needs of the user.
• Search: Use of AI to enhance search capabilities as 65 lakh misspellings and Hinglish searches happen on our platform every month. By comprehensively understanding Indian languages, including common misspellings and even ‘Hinglish’ or other multilingual search, we ensure precise search results, simplifying the process for users to find what they need effortlessly.
• Catalog management: AI-powered image editing tools have greatly streamlined and enhanced the cataloging capability for our users. From seamless background removal to automated cropping, these features have made product cataloging more efficient, elevating the visual appeal of listings and driving higher engagement.

When IndiaMART considers utilizing new AI, what is your number one consideration after all?

CX (customer experience) is the only consideration/ area where we would continue to invest. This includes all elements of technology innovation at IndiaMART; by CX we mean, improving search capabilities, redefining the value we offer customers (integrating WhatsApp, ease of use with LMS, evaluate security solutions for advanced threat detections, be safe with data, etc.) and when it comes to AI integrations, first and foremost, we need to validate how their solutions will make our customer's lives easier than today.

By bringing the customer into focus in every process, we are able to identify customer problems based off their feedback and journey on our platform, we simplify problems to see how we can design processes to address these, and then use technology to re-engineer existing processes to better serve customer experiences.

Is IndiaMART using AI to make searching for products easier? If so, how?

We are leveraging AI and machine learning to improve search relevance on the platform, provide personalized recommendations, and automate lead qualification. Close to 65 lakh misspellings and Hinglish searches happen on our platform every month. With a comprehensive understanding of Indian languages, including common misspellings and even ‘Hinglish’ or other multilingual searches, we ensure precise search results, simplifying the process for users to find what they need effortlessly.

Our search functionality also supports translation of Roman Hindi Queries, Identification of Entities in Search Queries (Attribute Extraction) and determining the Intent of the Query (e.g., Product or Company Type). By integrating generative AI, the aim is to simplify product discovery, enhance customer engagement, and optimize the matchmaking capabilities for businesses that visit the platform.

Does IndiaMART use AI to help with any tasks within the company itself?

Internal efficiency and optimization of organizational processes form a core part of my role at IndiaMART.

We are a 6000+ employee organisation spread across more than 28+ geographical locations in India. It is not humanly possible to run an organization this big without the integration of AI within our processes. Our robust CRM system is heavily incorporated with AI at multiple touchpoints, thus helping employees to engage with departments across cities, across offices seamlessly, for any resolutions.

Our sales and servicing team deals with the sellers daily via emails, calls, and physical connects. To ensure faster servicing and documentation, AI tools are integrated for transcription & summary of the conversations, thus ensuring utmost transparency with the customer.

In addition to this, the internal team also uses AI for relevant category mapping of seller products and supports them in case of any errors for correction. It is also used to propose possible products which the seller can add and help them create a robust catalogue, for more relevant buyer connections.

Are there new security risks because IndiaMART uses AI? What are you doing about them?

As the volume of data fed into systems increases, the concerns around data security are bound to increase. The potential risks include an increase in fraudulent activities and rising cases of data breaches. To address the same, we continually invest in evolving our practices to thwart potential data breaches, staying ahead of emerging threats. This includes:

• Advanced Threat Detection: Our strong encryption protocols and systems, such as integrated web application firewalls, a robust monitoring system for potential breaches, and other tools ensure overall data security. Routine security audits and penetration testing proactively identify vulnerabilities in our systems and promptly address these weaknesses. This helps in mitigating the risk of exploitation by cyber attackers, bolstering the resilience of our digital infrastructure.
• Robust security infrastructure: Our teams maintain a multi-layered defense strategy to protect the vast amount of customer data.  This includes following strict access control policies with role-based permissions and multi-factor authentication to limit data access to authorized personnel only.
• Incident Response and Disaster Recovery: In the event of a security breach, IndiaMART has a well-defined incident response plan to mitigate the impact and restore normal operations promptly. Furthermore, regular data backup procedures and a comprehensive disaster recovery plan are implemented to minimize downtime and data loss, ensuring business continuity even in the face of cyber attacks.
• Certifications & compliance:
  • Beyond the technical safeguards, we are fully compliant with relevant data protection regulations such as GDPR and ISO 27001, which ensures that our customers’ data is handled responsibly and securely.
  • IndiaMART holds several active certifications, including ISO 27001 for Information Security Management System, ISO 22301 for Business Continuity Management System, ISO 27701 for Privacy Information Management System, ISO 31000 for Enterprise Risk Management, and ISO 12207 for Software and System Engineering.

Is IndiaMART using AI to help protect itself from cyberattacks? How?

The increased sophistication of cyberattacks in the e-commerce sector makes it mandatory for any company to have a robust security infrastructure in place. At IndiaMART, we follow a multi-cloud strategy to distribute our workloads across our cloud partners. This prevents any bad actors at the initial stage itself. In addition to this, we have employed strong encryption protocols and systems in place, such as integrated web application firewalls, robust monitoring systems for potential breaches, continuous authentication, and strict access controls to create a secure and resilient platform for our users.

We have implemented AI-Driven Fraud Prevention & Real-Time Monitoring to identify and mitigate fraud, phishing, and account takeovers. In addition, we also use real-time transaction monitoring that helps in detecting suspicious activities and preventing fraudulent business engagements. The real-time transaction monitoring implemented within our systems also helps in detecting suspicious activities and preventing fraudulent business engagements.

Do you teach your employees about the security of using AI tools?

As the company embraces technology and explores solutions that can work better for our users, this creates opportunities for learning and growth within the organization, while also raising concerns around mandatory security training, as employees are usually the first line of defense and victims for any organization. We conduct comprehensive cybersecurity training for our workforce to educate them about potential threats such as phishing attacks, malware, and social engineering tactics. This ensures that our workforce remains vigilant and equipped to recognize and respond to security threats effectively.

Usage Guidelines for Public AI Tools: As AI tools like ChatGPT, Copilot, or code generators become popular, we’ve clear policies and internal awareness sessions in place to spread awareness around what data can and cannot be shared with third-party platforms. The goal is to avoid inadvertent data leakage or IP exposure.

AI Governance in Practice: We are also working toward establishing an internal AI Governance Framework that includes not just technical controls, but also ethical standards, usage boundaries, and audit mechanisms to ensure AI is used safely and responsibly across the organization.

How do you decide when to use new AI and when to focus more on security?

At IndiaMART, innovation and security go hand in hand - they are not mutually exclusive but complementary. We adopt a balanced, value-driven approach when deciding between deploying new AI capabilities and doubling down on security. Our decision framework is based on three core principles.

Business Impact First: We assess whether the AI initiative directly enhances user experience, operational efficiency, or marketplace trust. For instance, when AI improves lead quality, automates support interactions, or delivers personalized experiences, it moves to the top of our roadmap-but always with privacy and safety built in from day one.

Security by Design: Every new AI feature goes through rigorous checks for data privacy, access control, explainability, and ethical usage. In highly sensitive areas like customer data handling or financial transactions, we prioritize reinforcing security architecture before layering AI-driven intelligence.

Risk-Reward Alignment: If a new AI initiative introduces potential vulnerabilities or increases our attack surface, we delay or sandbox its deployment until the right security posture is established. Conversely, if security risks are emerging - be it from fraud, spam, or data misuse, we pause new experimentation to strengthen our digital defenses.

AI acts as an enabler of better security (e.g., fraud detection, spam filtering) and security at IndiaMART.. The balance is dynamic, but our north star is always long-term trust and value for users.

What do you think will be the biggest change in how AI and security work together at IndiaMART in the future?

Deep convergence of AI and security is an opportunity, rather than a challenge, where AI is no longer just an innovation layer, but an integral part of the cyber defense fabric. As we scale our platform to support increasing buyer-seller interactions daily, manual or rule-based systems alone won’t suffice. We believe that AI will increasingly become our first line of defense, enabling real-time anomaly detection, behavioral threat analytics, and dynamic risk scoring. For instance, instead of flagging spam or fraud reactively, we’ll have AI models that proactively learn evolving attack patterns, detect subtle signals of abuse, and auto-mitigate threats before they impact users.

Additionally, AI governance and ethical usage will also become a key security priority. As we adopt large language models and generative AI in lead verification, communication, and automation, protecting against model misuse, hallucination, and data leakage will be just as important as traditional security controls. In essence, we’re moving toward a future where AI secures the platform and the platform secures the AI. This symbiosis will redefine how we think about risk, trust, and resilience in a digital marketplace like IndiaMART.