FMCG and QSR: Shelf is the new research lab

Across FMCG and QSR, the new product launch is turning into a live experiment, where the customer's next order, not a focus group, decides what earns a permanent place on the shelf.

There was a time when a new product was treated almost like a verdict, arrived at after months of boardroom debate, taste panels and consumer surveys, then delivered to the market fully formed and rarely revisited. That certainty is fading fast. In its place has arrived something closer to a live experiment, a hypothesis tested in the real world, refined in public, and judged not by what people say they want in a research room but by what they actually choose to buy again.

The instinct driving this shift is not new to the technology world, where building a rough version of a product, putting it in front of real users and improving it based on their behaviour has been standard practice for years. What is new is how comfortably that instinct is now travelling into aisles of ice cream, donuts and packaged yoghurt, categories once defined by long gestation periods, cautious testing and capital-heavy national rollouts that could take a year or more to plan.

The result is a quiet but consequential change in how India's food and beverage companies define success. Distribution used to be the finish line, the moment a brand could declare a launch complete. Increasingly, it is only the starting gun. The real test arrives weeks later, in the form of a second purchase, a reorder, or a customer returning without being asked to. That single behaviour, repeated across thousands of consumers, has begun to matter more than any survey ever could.

The shelf becomes the lab

Quick commerce, e-commerce and direct-to-consumer platforms have quietly rewritten the economics of trying something new. A brand no longer needs to commit to a national rollout to find out whether an idea works. It can test a flavour in a handful of pin codes, offer it through a single platform, or restrict it to one city, and read the results in near real time through ratings, reviews and repeat orders.

"There has been a fundamental shift in how FMCG companies approach innovation over the last five to seven years, driven largely by the rise of e-commerce, quick commerce and D2C channels," says Ritesh Gauba, CEO at Epigamia. "Today, brands can launch products in limited quantities, target specific geographies, pin codes or consumer cohorts, or choose to partner with select e-commerce platforms instead of rolling out nationally from day one. This allows companies to take a launch, learn and scale approach by testing products with the right consumers before making larger investments."

What has changed underneath this is the speed of the feedback loop itself. A decade ago, understanding whether a product had genuine appeal meant waiting for quarterly sales reports or commissioning fresh research. Today, that same understanding arrives through star ratings, comment sections and reorder patterns within days of a launch, allowing companies to make faster, better informed calls on whether an idea deserves a wider audience or a quiet exit.

When the reorder becomes the real verdict

Nowhere is this metric shift more visible than in quick service restaurants, where limited-time offers have quietly become a testing ground rather than a marketing gimmick. For a food business, that philosophy is explicit. "The shelf is our research lab and the customer's next visit is our data point," says Tarak Bhattacharya, CEO of Mad Over Donuts. "Kit Kat and Kunafa didn't make it to our permanent menu because a focus group approved them, they made it because people kept coming back. That's the only research that's never wrong."

Bhattacharya points to a change in what the brand even chooses to measure. Opening day footfall, he notes, only proves curiosity, not commitment, which is why the company has shifted its attention from day one queues to what happens in the second and third week of a launch, whether a product like the Dubai Chocolate Shake or the Cereal Donut is still being sought out once the initial buzz has settled. Even culturally loud collaborations are now judged less by the noise they generate and more by whether that noise converts into people coming back for it once the novelty fades.

That thinking finds an echo at a ready to serve products like ice creams, where unconventional flavours such as Smoky Charcoal, and formats built on camel, goat and donkey milk, are treated as conversations rather than conclusions, as reflected by Dr Suhas B. Shetty, Founder and CEO, Iceberg Organic Ice Creams. "Consumer preferences are evolving much faster than before, and brands need to adapt accordingly," explains him. "I believe launches today should be seen as the beginning of a conversation rather than the final product. At the end of the day, repeat purchases, customer feedback and organic word-of-mouth tell you far more than traditional research ever can."

Why food cannot fully copy software

Yet the comparison to software has its limits, and none of the three leaders pretend otherwise. A buggy app can be patched overnight after release. A donut or an ice cream cannot be quietly rewritten once it has reached a customer's hands. "Taste has to be right from day one, and every experiment carries real production cost, not just testing cost," is how Bhattacharya frames the difference, adding that the brand is not tweaking a product after launch so much as using the launch itself to decide what deserves a permanent place on the menu.

Gauba draws a similar line around scale and price point. For niche or premium products, he says, trials and repeat purchases genuinely indicate success, but for mass-market products sold at five or ten rupees, distribution remains the primary driver regardless of how the newer feedback loops evolve. Shetty adds that sourcing, manufacturing and quality consistency mean experimentation at Iceberg is always balanced against operational discipline, since a controlled test market still has to deliver the same product reliably at scale.

The new scorecard for innovation

What emerges from these three very different businesses is not a single template but a shared instinct: the customer's second decision matters more than their first, and a limited launch is now a legitimate way to learn before committing serious capital to a national bet. For an industry built on wide, expensive rollouts, that is a meaningfully different starting point.

None of this replaces judgment, sourcing discipline or the hard economics of manufacturing at scale. But it does suggest that the old order of research, then launch, is steadily giving way to a newer one, where the launch is itself the research, and the shelf, not the survey, delivers the final word.