RasoiSense

The startup set out to build a PetPooja alternative — but talking to restaurant owners in Hyderabad surfaced a sharper problem than "operations software." Owners weren't just frustrated by clunky order management. They were losing money quietly, every week, to food waste and guesswork.

"We need to be prepared if the stock runs out — we don't want to over-buy stuff, and we want everything to be fresh."

That single sentence contains two opposing risks restaurant owners juggle constantly with no real tooling to help: over-buy, and you waste money and freshness; under-buy, and you run out mid-service and lose the sale entirely.

Existing tools like PetPooja focused on transactions — orders in, orders out. None of them helped an owner see ahead of a problem. RasoiSense's opportunity wasn't to build a better version of the same tool. It was to build what PetPooja never tried to be: a system that helps restaurants see stock foresight coming, before it costs them money.

PetPooja is the dominant player in this market — a comprehensive system covering POS, billing, CRM, loyalty, multi-outlet reporting, and inventory tracking with low-stock alerts, among dozens of other features. It's built to be a restaurant's entire operational backbone, and for larger or multi-location businesses, that comprehensiveness is the appeal.

For an independently-owned restaurant, it's also the friction. Inventory foresight exists in PetPooja, but it's one module buried inside a system built for a different scale of business — the kind of setup that assumes someone on staff has the time and technical patience to configure it. What owners told us they actually wanted wasn't another report to check. It was an answer to one question, front and center: what am I about to run out of, and what am I about to over-order?

RasoiSense's bet isn't "more features than PetPooja." It's narrower on purpose — built around that one decision, for owners who don't have a dedicated person to run the software.

Before any screens, I mapped the three roles who'd actually touch RasoiSense day to day — Host/Waiter, Kitchen staff, and Manager — in FigJam, tracing how information about a single order moves between them in a real shift.

The insight that shaped everything downstream: these three roles don't need a shared view of the same data. They need almost entirely different views of it, at different moments, under different pressure. A waiter needs speed mid-service. Kitchen staff need urgency and legibility from across a room, often with their hands full. A manager needs the opposite — patterns over time, not real-time noise.

One design principle governed every screen from here on: Hick's Law — the more choices someone faces, the longer it takes them to decide. For staff who are moving fast, distracted, or new to the job, every extra option on screen is a cost. So the standard I held every screen to was ruthless reduction: if a screen could function with fewer choices visible at once, it had to.

Why Host/Waiter first. The three-stage build order — Host/Waiter MVP, then Kitchen sync, then Manager dashboard — wasn't arbitrary. Order flow is the one workflow every restaurant runs constantly, regardless of size or type, and it's the fastest way to prove the core product works in a real shift before layering on more complex pieces. Building all three simultaneously would have meant testing nothing in production while carrying the most risk. Shipping the simplest, highest-frequency workflow first meant real feedback, faster — and a foundation the other two stages could build on rather than guess at.

Designing for a chef who can't look away. The kitchen board was the clearest test of Hick's Law in practice. Chefs are already at capacity — managing multiple dishes, timing, and heat, often without a free hand. The board couldn't ask them to parse anything beyond what actually matters in that second: what's cooking, and how much capacity they have left. Every other piece of information — order history, customer notes, anything that wasn't immediately actionable — was deliberately kept off that screen. Capacity and active dishes needed to be readable at a glance, not read.

What we chose not to build yet. Early on, adding delivery aggregator integrations — Swiggy, Zomato — came up as an obvious feature. We decided against it, for now. The reasoning: RasoiSense needed to prove staff could actually adopt the core workflow first, before adding the complexity of a third-party delivery layer on top of a tool they were still learning. Aggregator integration is a real feature on the roadmap — but shipping it alongside an unproven MVP would have meant testing two unknowns at once instead of one. It's staying deliberately out of scope until the foundation is validated.

The palette — terracotta, leaf green, amber, on a warm off-white base — wasn't a food-app default. Two things drove it.

Practically: earth tones are easier on the eyes over a long shift. Chefs are staring at a kitchen board for hours under kitchen lighting; managers are looking at dashboards all day. Sharp, saturated colors (the red/orange most food apps default to) fatigue faster than warm, muted ones.

Culturally: Rasoi means "kitchen" in Hindi — and in traditional Indian households, the kitchen floor and walls were often finished in terracotta, bordered with white line detailing. The palette isn't decoration on top of the name. It's the name, translated into the interface. Bricolage Grotesque carries that same warmth into the typography — approachable and a little handmade, rather than sterile SaaS-default sans.

This project reinforced something I keep relearning — restraint is a design decision, not a compromise. Every screen in RasoiSense got stronger by removing something, not adding it: fewer choices for a distracted waiter, less noise for a chef mid-shift, one clear number instead of a dashboard for a manager who just needs to know what to buy. The hardest, most senior part of the job wasn't designing the insights screen. It was saying no to Swiggy and Zomato integration until the core workflow earned the right to get more complex.