E-commerce marketplace Meesho has introduced “Vaani”, a generative artificial intelligence-powered conversational voice shopping assistant aimed at simplifying digital commerce for India’s next wave of internet users.
The company stated that the assistant has been designed to cater to the next 500 million users who continue to rely primarily on offline shopping and often find conventional e-commerce interfaces—centred on typing, filters and structured search—difficult to navigate. It added that the assistant seeks to replicate the natural, conversational style of offline shopping within a digital environment.
In physical retail settings, consumers typically describe their requirements, ask questions, compare options and refine their preferences through dialogue before making a purchase. Meesho noted that such behaviour has not translated effectively to online platforms, particularly among users in tier-2 cities and smaller markets.
Vaani enables users to interact using natural speech, allowing them to ask follow-up questions and progressively narrow down choices through conversation, rather than relying on typed queries or multiple filters. The assistant is designed to support the entire shopping journey, from product discovery to final purchase, by interpreting user intent, surfacing relevant options, providing contextual information such as reviews and product details, and assisting with payment and delivery processes.
Sanjeev Kumar, Co-founder and Chief Technology Officer at Meesho, stated that artificial intelligence had long been central to the company’s approach to building scalable solutions. He noted that with over 251 million annual transacting users, the company possessed a strong understanding of consumer behaviour in India, adding that Vaani integrates conversational AI with the platform’s commerce intelligence to guide users seamlessly from discovery to purchase. He further stated that the launch represented a significant step towards making e-commerce more accessible and intuitive.
The system is built on a multi-layered AI architecture tailored for real-world commerce applications. It incorporates edge computing for speech recognition and synthesis, enabling faster response times while optimising operational costs. A multi-agent framework allows the assistant to manage complex, multi-step interactions, while models have been fine-tuned to account for regional language nuances. The assistant also features multi-modal capabilities, enabling it to interpret both voice inputs and on-screen context.
According to the company, early adoption metrics indicate strong user engagement. Within the first month of launch, more than 1.5 million users interacted with Vaani, with repeat usage suggesting the beginnings of habitual adoption. Meesho reported that 79% of users found voice interactions to simplify shopping, while 94% described the experience as intuitive. Approximately 62% indicated trust in the assistant for completing transactions.
The company added that users engaging with Vaani demonstrated 22% higher conversion rates, progressing more quickly from product discovery to purchase, alongside lower rates of returns and cancellations.
India currently has an estimated 850 million internet users, of whom only 250–300 million transact online regularly, highlighting significant potential for further growth in digital commerce adoption.
