AI-Driven Festive Shopping Era is Here, But What Else to Expect?

Festive season has been a litmus test for internet companies which have bet big on AI deployment across the workflows. 

The AI deployment here is not limited to just customer support but for discovery and personalisation among others. More on this later.

We have previously talked at length about the deployment of AI in place of human customer support. Just to recap, several top tech companies, including the likes of Zomato and Paytm, have restructured their workforce to align with AI. 

READ: AI Leads To Job Cuts In Customer Support: But Will Push For Agentless Setup Hold?

It also appears that consumers are too embracing this AI transition. According to NIQ’s Agentic Commerce India 2026 study, 92% of consumers across India’s top metros had already used AI while shopping this year, and 96% said they were open to AI-powered shopping tools.

According to InMobi Advertising’s 2026 festive season report, four in five Indian shoppers expressed interest in using AI-powered shopping assistants. The report also found that 89% of consumers showed interest in smart search suggestions and 81% in virtual product try-ons. 

Google’s 2026 festive consumer research found that Indian shoppers used generative AI chatbots an average of 7.5 times during their Navratri and Diwali shopping journeys. These findings indicate that consumers are becoming comfortable using AI as an active decision-support layer rather than simply a source of information.

According to a report, Meesho’s inaugural day of its sale saw the majority of orders coming in via AI-powered discovery and personalisation. Amazon, the report added, saw its AI-powered Rufus doing the heavy duty work during its annual sale.  

Additional layers

AI has become an increasingly active participant in the festive commerce cycle, influencing how consumers discover products, evaluate options, interact with brands and complete purchases. Its role has expanded considerably beyond recommendations and automated customer support. Businesses are now using AI to interpret consumer intent, personalise interactions, anticipate demand and manage large volumes of customer conversations across multiple channels.

Mashhood Raza, Founder and CEO at Voxforce AI, highlights another significant development is the movement from generative AI to agentic AI. Traditional automation responds to predefined instructions, while agentic systems can interpret objectives, determine the next steps, access relevant information and execute tasks within defined permissions. During high-demand periods such as the festive season, this capability becomes particularly valuable. 

“An AI agent can understand a customer’s requirements, retrieve relevant product or service information, address follow-up questions and facilitate the next stage of the interaction without requiring a human employee to manage every step,” Raza added. 

Earlier this year, GoKwik and PayU launched India’s first multi-brand D2C commerce experience built directly into ChatGPT, letting shoppers discover and check out from early D2C brands like Hyphen, Beardo and Kilrr inside a single AI conversation. 

“… With hundreds more expected to join through the GoKwik ecosystem in a phased rollout — reaching ChatGPT’s 700 million weekly users. That’s not a discovery feature. That’s a checkout running inside someone else’s AI assistant, settled over UPI, cards, net banking and wallets, with brands keeping full control of their catalogue and customer data,” Chirag Taneja, Co-founder & CEO, GoKwik tells Entrepreneur India. 

Taneja added that it has seen a similar shift inside their own stack. Citing the example of its Kwik Ads, which is based on purchase intelligence from over 200 million shoppers, Taneja said, it runs signal enrichment, creative research, campaign benchmarking and spend analysis the way a performance marketer would work a campaign, not the way a static dashboard reports one. 

“Brands on it – Personal Touch Skincare, Pee Safe, Bombay Shaving Company, What’s Up Wellness, and Saada among the early cohort have seen a 25% improvement in return on ad spend and a 15% reduction in customer acquisition cost; one brand saw a 7% lift in prepaid orders purely from better real-time signal tracking,” he added. 

A temporary spike?

This year’s festive sales have been remarkable, even though they missed the experts’ forecasts. 

According to a report, ecommerce and online retailers scooped INR 1.40 lakh crore in gross merchandise value (GMV) during the festive month. The volume shows that AI can handle scale. But these figures will cool down as the festive season concludes. 

It’s quite similar to the spike in the US during the online holiday sales. For instance, Adobe is projecting the US online holiday sales to go past USD 275.1 billion during November and December period. 

But what happens after the spike phase is over?

Ashutosh Prakash Singh, Co-founder and CEO  RevRag AI, tells Entrepreneur India that a more consequential development after the festive season will be the way businesses use AI to improve the economics of their operations. Seasonal demand creates short-term pressure on acquisition, servicing and transaction volumes. 

However, the underlying business priorities remain constant throughout the year: acquiring the right customers, improving conversion rates, reducing acquisition costs, increasing customer lifetime value and maintaining portfolio quality.

Financial services companies have considerable scope to apply AI across these priorities. One important area is intelligent customer acquisition. Rather than treating every lead as having equal value, AI can help institutions analyse customer profiles, identify relevant product requirements and prioritise follow-ups based on the likelihood of conversion. This can improve the productivity of sales teams and reduce expenditure on low-intent leads.

Another area is the economics of customer servicing. 

“Banks, NBFCs, wealth management platforms and digital lending businesses manage large customer portfolios with varying requirements. AI can help segment these portfolios, identify customers who may need intervention and determine the most relevant communication or service action. This is particularly useful in lending, where timely engagement can influence application completion, repayment behaviour and customer retention,” he added.

Devyani Gupta, Founder & CEO, Arrowhead tells Entrepreneur India that they see three shifts that make this permanent. First, assisted commerce is becoming the default for Bharat. About two-thirds of this season’s shoppers came from Tier II cities and beyond, and first-time shoppers from these towns roughly doubled. For someone buying online and taking credit for the first time, a voice that explains the next step in Tamil, Telugu or Hindi isn’t a convenience. It’s what builds trust. Second, AI is moving from a cost line to a revenue line. 

“The question in the boardroom used to be how many calls AI can handle; now it is how much revenue AI can recover. Our partner extended the programme beyond the sale and widened it from one stage of the journey to the entire funnel. And third, festive credit has an afterlife. With BNPL and checkout EMI volumes set to reach USD 30.5 billion in 2026, every financed purchase starts a relationship that needs onboarding, first-EMI reminders and servicing,” Gupta said. 

“The way we approach it is simple: observe, segment, replicate, compound. A customer who intends to buy and one who doesn’t should never get the same conversation, so we segmented stalled shoppers into five behavioural cohorts. Someone who has already bought gets thanked and released in seconds. Someone stuck on a mandate or an OTP gets walked through that exact step, live. Someone who wants a different product becomes new demand, not a lost lead. Someone who needs time gets called back when they choose, and about one in seven booked their own slot. And someone who isn’t ready is respected, with strict contact limits and no pressure. Within three days of applying this, positive outcomes rose 37%, and fixes moved from feedback to production in under 24 hours. In a sale that lasts under two weeks, learning speed is the strategy.”

AI Cranked Up

We’re expecting a much deeper AI integration in quick commerce in the near future, regardless of the seasonality of the festive season. 

Experts believe that one of the biggest gaps in current AI deployment is the limited use of customer intelligence beyond personalisation and recommendations. Consumer-facing internet companies have access to substantial amounts of behavioural and transactional data, yet much of this information remains underutilised when it comes to understanding a customer’s underlying needs, financial capacity and long-term value.

That said, in financial services, this limitation can have direct commercial consequences. A customer may show interest in a financial product but may not be ready to proceed because of affordability concerns, uncertainty around product terms or a mismatch between the offering and the customer’s requirements. Simply increasing the frequency of communication will not resolve these issues. Businesses need systems that can interpret why customers are dropping out of a journey and identify what intervention could genuinely help them move forward.

According to Singh, data readiness is another major constraint. Dun & Bradstreet’s 2026 AI Momentum Survey found that only 4% of surveyed Indian businesses considered their data fully ready for AI. This points to a fundamental problem. AI systems are only as useful as the information available to them. Inconsistent customer records, fragmented transaction histories and limited integration between business systems can restrict the quality of insights and recommendations.

There is also a need for greater specialisation. A general-purpose AI system cannot be expected to address the different requirements of digital lending, wealth management, insurance, investment distribution and banking with the same level of accuracy. Each category has its own customer behaviour, regulatory considerations, risk parameters and decision-making processes. Financial AI needs to reflect these distinctions.

Gokwik’s Taneja adds that almost everything consumers have actually touched this festive season sits in discovery search, recommendations, try-ons, and styling. Execution – checkout, payments, and what happens after the order is placed is still catching up, and it’s precisely the layer most exposed to drop-off, failed payments and support load during a sale.

“Agentic commerce today runs almost entirely through a small number of surfaces, ChatGPT chief among them. We’d expect the next round of festive season to see the same commerce layer extend to other AI assistants and messaging surfaces, rather than remaining a single-platform story and we’re building GoKwik’s own roadmap on that assumption, not as a bet on any one assistant staying dominant,” he continued.

To sum it up, it’s safe to predict that in the next festive season, we can expect diversification across at least three key areas. First, AI agents will move from answering questions to completing multi-step customer workflows. Second, multimodal engagement will become more prominent, with voice, video and text working together to create more natural interactions. Third, businesses will increasingly deploy specialised agents across the customer lifecycle, covering discovery, sales, onboarding, support, retention and re-engagement.