When Magicpin launched its AI assistant, Vera, this week, the headline wrote itself: a hyperlocal platform rolls out an AI co-pilot for small restaurants and retailers. Backed by a $1 million investment, 100,000 merchants are already onboarded during the trial, and with early results showing up to a 3x increase in customer actions, Vera becomes credible.
But the more interesting question is not what Vera does. It is why something like Vera needed to exist at all.
India has 7.5 million restaurants and food outlets, most of them single-owner operations running on margins thin enough that a bad week of footfall can threaten the month. For years, these businesses were told that going digital was the answer — get on Swiggy, get on Zomato, get discovered. Many of them did exactly that. And many of them are still waiting for the part where it actually worked in their favour. It is about what happens to these businesses when the digital economy that was supposed to democratise commerce instead created a new set of gatekeepers. And it is about whether AI — the same technology that has disrupted everything else — can hand merchants the agency that food delivery aggregators were never designed to give them.
The Aggregator Bargain
To understand why Vera matters, you have to first understand the trap that came before it. When Swiggy and Zomato scaled through the 2010s, they offered small restaurants something genuinely valuable: overnight access to thousands of customers they could never have reached on their own. The trade-off felt reasonable at the time. A listing, some visibility, a stream of orders. In exchange: a commission.
The commissions have not stayed reasonable. Zomato and Swiggy typically charge commissions of 16–30% per order. That figure, though, is just the starting point. Layer in the GST on platform fees, payment gateway charges, packaging costs, and the increasingly prevalent “visibility assurance” or priority listing fees — and a restaurant that appears to be selling a ₹300 dish online may be walking away with well under ₹200. One breakdown from Spice Advisors puts it starkly: some restaurants lose over 35% of revenue before raw materials, packaging, rent, or labour are even accounted for.
And that is before you get to the deeper problem: the data. A restaurant selling a hundred orders a day on Zomato has no meaningful relationship with any of those customers. It does not know who they are, where they live, what they ordered before, or why they stopped ordering. That information lives on the platform. The restaurant itself is, in the language of digital commerce, a supplier to a marketplace — not a brand building its own audience.
In mid-2025, Zomato had introduced a new long-distance delivery fee — charged to restaurants, not customers — on top of already-rising commissions. Restaurant partners told the publication they were being auto-enrolled in ad campaigns and billed without giving explicit consent, with ad spend climbing week to week even when order volumes stayed flat. The National Restaurant Association of India (NRAI) took up the matter with Zomato CEO Deepinder Goyal directly, with the association describing the platform’s initial posture as “aggressive.” Goyal later assured the NRAI that coercive tactics would not continue — but the underlying model did not change. Restaurants that rely on aggregators for their visibility have very little room to negotiate, because walking away from the platform often means walking away from their only meaningful source of online orders.
The Visibility Gap Nobody Solved
Here is what makes the aggregator problem particularly stubborn: the alternative is genuinely hard. Getting discovered outside of Swiggy and Zomato means owning your Google Business Profile, generating reviews, showing up in local search, managing responses, and increasingly — as search behaviour shifts — appearing in AI-powered discovery channels like ChatGPT and similar tools.
For a restaurant owner running a 12-hour operation with a kitchen team to manage and margins to protect, this is not a marketing strategy. It is a full-time job on top of the actual job. India’s AI in food industry report via IndiaAI.gov.in confirms that while AI adoption in food retail is accelerating, the benefits have disproportionately flowed to larger, better-resourced operators — chains, cloud kitchen brands, and QSR groups with dedicated digital teams. The long tail of independent restaurants, which Aaron Allen & Associates estimates makes up nearly two-thirds of all restaurant units in India, has been largely left out.
This is the gap that Magicpin is positioning Vera to fill. The assistant is designed to handle Google listing optimisation, automated review responses, AI-channel visibility, and lead conversion — functions that a restaurant would otherwise need to either ignore, outsource expensively, or attempt manually. India’s food services market is valued at $85.19 billion in 2025, growing at over 10% CAGR. The merchants at the base of that market have been largely spectators to the digital infrastructure being built around them. Vera is, at least in intent, an attempt to change that.
What Vera Is Actually Claiming
Magicpin CEO and co-founder Anshoo Sharma has been direct about what the company expects from this. In a statement accompanying the launch, he described Vera as more than a feature addition:
“Vera acts as a growth engine that not only manages a merchant’s digital presence but also drives real business outcomes.”
The company reports early results showing 1.5 to 2 times uplift in visibility and up to three times increase in customer actions among merchants in the trial cohort — which, notably, already comprises over 100,000 outlets across India. That is not a small sample. The $1 million investment in Magicpin’s AI stack signals that this is not a feature release; it is the beginning of a new business vertical, one Sharma believes could be a multi-billion dollar opportunity over the next five to seven years.
There is also a timing detail worth noting. Magicpin did not wait for a polished launch window. The company advanced Vera’s rollout in response to the LPG shortage that disrupted food delivery operations earlier this year — using the AI layer to provide real-time operational support to partner restaurants during a period of genuine stress. The crisis became a forcing function, and the accelerated deployment means Vera has already been stress-tested at scale before its formal announcement.
Why This Is Not Just Another SaaS Play
There is an instinct, when covering AI product launches in India, to file them under “startup claims” and move on. A lot of merchant enablement tools have come and gone. What is different about this moment — and about what Magicpin is attempting — is the convergence of three things happening simultaneously.
First, the search itself is changing. Google’s AI Overviews and conversational AI tools like ChatGPT are beginning to reshape how consumers discover local businesses. A restaurant that is not optimised for AI-driven discovery does not just miss a new channel — it risks disappearing from the primary surface where decisions get made. For small merchants who have never managed a Google Business listing properly, this shift is not just an inconvenience. It is an existential threat to organic discovery.
Second, Rapido‘s entry into food delivery with commission rates of 8–15% — roughly half of what Zomato and Swiggy charge — is evidence that the aggregator model is under competitive pressure for the first time in years. That pressure creates an opening for tools that help merchants build independent demand, rather than remaining entirely dependent on whichever platform offers the most traffic at any given moment.
Third, and perhaps most importantly, the cost of AI infrastructure has dropped to a point where a ₹300 dish seller in Patiala can theoretically access the same quality of digital marketing automation as a well-funded QSR chain. That was not true two years ago. The tools simply did not exist at a price point that made sense for micro-merchants.
The Question That Remains
None of this means Vera will work as advertised, or that Magicpin will succeed in building a sustainable SaaS business on top of a merchant base that has historically been resistant to paying for software subscriptions. India’s restaurant sector is notoriously hard to monetise at the long tail. As Restaurant Coach’s analysis of the aggregator landscape notes, the real challenge is not convincing restaurant owners that digital presence matters — most of them already know it — but building something simple enough that a kitchen-first operator can actually use it without a dedicated marketing person on staff.
That is where Vera’s “always-on co-pilot” framing is either its strongest asset or its biggest risk. If the interface genuinely requires minimal intervention — handling reviews, updating listings, converting queries — it could become as habitual as a POS system. If it demands configuration, training time, or active management, adoption at the long tail will stall, regardless of how good the underlying AI is.
Anshoo Sharma has previously described Magicpin’s philosophy in terms that feel relevant here. In an earlier interview, he captured the fundamental problem with India’s local merchant economy: “The local merchants of our country, who are the backbone, are unable to advertise online because they can’t pay for views.” Vera is, in essence, Magicpin’s answer to the same problem — but in the era of agentic AI rather than performance marketing.
Anshoo Sharma has built Magicpin around a single conviction: that India’s local merchants are the backbone of its retail economy but have always lacked the tools to compete online. The platform started as a hyperlocal rewards app in 2015, expanded through ONDC, and has now crossed 15 million active users — but the merchant side of the equation has always been its core bet. Vera is, in essence, the most direct expression of that bet yet. Rather than just routing orders to restaurants, Magicpin is now trying to give them the infrastructure to generate their own demand — in the era of agentic AI rather than performance marketing.
Whether it succeeds or not, the launch points at a real tension that has been sitting underneath India’s food delivery economy for years. The aggregators built the demand. They never gave the merchants the tools to own it. If AI can change that equation — even partially, even imperfectly — it would represent a more meaningful shift than a hundred commission percentage points renegotiated at the NRAI table.
India has 7.5 million food businesses. Most of them are invisible on the internet. That is not a technology problem anymore. It is a distribution problem — and someone just decided to make it their business to solve it.
