Consistency Is the Product: What a Bihar Badi Brand Taught Me About Scaling Fragile Systems

Every architect I respect can spend an hour on uptime. Very few can tell you how they’d guarantee output quality when the production unit is unpredictable – sunlight, weather, human hands – and the promise to the buyer is “same taste, every single month, anywhere in India.” That is the harder problem. And someone in Patna is quietly solving it.

Shital Kumari lost her mother, went looking for the badi that tasted like home, and found nothing on the shelf she could trust. So she built GomatiRoots – funded partly by state schemes, a small loan, and a Rs 1 lakh grant through YourStory’s Shuru Kar initiative. First-year revenue: Rs 32 lakh. Fifty-plus stores, three marketplaces, a D2C site. Strip away the sentiment and what remains is a masterclass in systems thinking.

The Constraint Nobody Questioned

Badi is sun-dried. For generations, that meant seasonal production, weather-dependent quality, and a product that stayed local. Everyone in the category accepted this as natural law.

She treated it as a defect.

That is the first discipline of a good architect: separate the constraints that are physics from the constraints that are merely inheritance. Most “that’s how it’s always been done” is not a law – it’s an unimplemented requirement. Once you accept that badi must be available year-round at consistent quality, the entire problem shifts from recipe to process: drying control, moisture specification, packaging, shelf life. The product didn’t change. The reliability engineering around it did.

Where the Architecture Actually Lives

Retail chains don’t buy badi. They buy a promise about the next delivery. In distributed systems we call this idempotency – same request, same result, every time. A household food item becoming a brand requires exactly that: idempotent output.

This is where most rural micro-enterprises collapse. Self-help groups are superb production nodes – proven skill, minimal capex, extraordinary rural reach. But nodes are non-deterministic. You cannot build a national category on goodwill and memory. You need a process contract at the node and a quality gate at the aggregation layer. Whoever standardises that layer – not the product, the process – owns the category.

There is also formalization debt: licences, packaging compliance, cold storage, distributor documentation. Founders treat this as paperwork. It isn’t. It is the substrate on which every future release depends. Defer it, and you’ve scheduled the failure of your own next version.

And note the capital story: roughly Rs 4–5 lakh of total input against Rs 32 lakh of revenue. A ~7x asset turn. That isn’t just grit – it’s architectural restraint. She built the smallest system capable of proving the thesis, not the biggest one she could finance.

The Northeast’s Opposite Asymmetry

We have the mirror image of this problem, and it should worry us.

The Northeast has secured GI tags – Lakadong turmeric, Naga chilli, Chak-Hao black rice. A GI tag is a namespace. A brand is the application layer that consumers actually invoke. We are world-class at building registries and celebrating them, and remarkably weak at the unglamorous layer above: consistency, packaging, shelf life, distribution, working capital.

Every NRLM cluster across Assam, Meghalaya, and Nagaland has production capacity. Almost none has made a brand promise. The gap was never skill. It was standardisation. When I sit with founders at STPI, that is the gap I point to first – while somebody else builds the application layer and captures the value of our namespaces.

What I’d Tell a Founder

• Consistency before scale. If output varies, growth merely accelerates distrust.
• Audit your inherited constraints. Half of them are hypotheses, not laws.
• Pay formalization debt early. It becomes the ceiling, silently.
• Measure repeat orders, not funding. A shopkeeper reordering is the only KPI that cannot be faked.

Every brand begins as a memory of somebody’s kitchen. The work of architecture is making sure that memory survives the scale. Hokar Nidar Shuru Kar – but build the process before you build the dream.

---

About the Author: Sanjeev Sarma is the Founder Director and Chief Software Architect at Webx Technologies. With a core focus on Generative AI integration, Cloud-Native Scalability, and Enterprise Software Architecture, he has spent over two decades driving digital transformation across Northeast India and beyond. Beyond his corporate leadership, Sanjeev is deeply invested in shaping the future of the IT industry. He serves as an Industry Expert on the Board of Studies for Assam Don Bosco University’s School of Technology, advises state technology committees, and actively mentors emerging tech startups at STPI. He brings a unique, dual perspective of high-level enterprise execution and future-ready academic curriculum development.