Quick Commerce in India didn’t start with Blinkit or Zepto; it was already being explored behind the scenes at companies like Flipkart during the early days of COVID. Gauri Sapna, part of Flipkart’s Supply Chain Design Team, was closely involved in shaping one of the company’s earliest attempts at rapid delivery.
She worked on several high-impact projects, including the company’s early Quick Commerce initiative. In this conversation, she shares practical insights from building dark store models, handling fresh produce challenges, and optimizing inventory without historical data.
You were part of Flipkart’s early Quick Commerce team. What was your role, and what did it involve?
I was part of the Supply Chain Design Team. We were responsible for building new supply chains from scratch — things like warehouse sizing, location strategy, picking processes, and last-mile models. During COVID, we worked on Flipkart’s second attempt at Quick Commerce — even before Blinkit and Zepto took off.
The goal was 90-minute delivery across a wide product range, from groceries to gadgets. That meant designing for speed and flexibility.
We built dark stores in urban areas — small warehouses of about 500 to 2,500 sq. ft. Since there were no playbooks in India at the time, we had to figure out everything from scratch.
One of the most effective techniques we used was random storage. Instead of fixed item locations, we allowed storage wherever space was available, following simple rules like keeping food and non-food separate. This helped us utilize space better and made operations more agile.
A lot of people hesitate to buy fruits and vegetables online. How did you deal with that?
That was one of the biggest pain points. The moment you pluck a fruit, it starts to deteriorate. In the early days, we had a lot of complaints — especially fresh.
We had to change the entire approach. We studied how produce behaves — things like ethylene interactions, ripening timelines, how bananas affect other fruits, that sort of thing.
There was also a big debate on air-conditioned storage. Some said vendors don’t use it — but they sell everything within a day. We needed to hold stock for 1–3 days, so storage mattered.
By improving packaging, segregating carefully, and managing ripening, we got much better at it. In fact, we ended up with one of the highest customer satisfaction scores in the fresh category.
Quick Commerce often lacks historical data. How did you manage inventory and demand forecasting?
Since the model was new, there wasn’t much historical data to rely on. We used a hub-and-spoke model. Perishables like dairy and bread were delivered directly to dark stores, but non-perishables were stored at larger warehouses and transferred to dark stores based on early demand signals.
This gave us better control, helped reduce wastage, and allowed more flexibility when it came to vendor negotiations.
Did you face any gender bias?
Warehouses were often in remote areas, which made things like transport and safety harder — especially for women. Even if there was a cab, you couldn’t always rely on it being safe or timely. And for late shifts, if a security guard wasn’t available, I had to leave, no matter how urgent the work was.
Also, there weren’t many women in senior roles. That makes it harder for women in junior roles to see themselves growing long-term in the supply chain.
What would you say to someone just starting out in the supply chain?
Try different areas — operations, design, planning — to understand how the whole system works. Don’t work in silos. And focus on stakeholder management. You’re always working across teams — it’s not just about solving the problem, but also about getting everyone aligned on the solution.
