Saahil GoelCo-founder & CEO

We started fourteen years ago in 2011 as Cart Rocket because we saw that Indian merchants had no easy way to build websites. I was a coder and building stores seemed like a great opportunity. However, we pivoted twice, almost died both times, and had to learn that the Indian market was completely different from the West.

Honestly, we did not validate the idea properly initially, which was a mistake. I advise founders to look for what customers are complaining about, rather than just building what they personally like. The best place to start is finding the intersection of your passion and customer complaints.

We launched using Magento, which is an open-source e-commerce platform. We called it Cart Rocket, sold it to enterprises, and acted as an agency. I pretended we had a big team and did all the coding myself. That agency revenue funded our first two engineers before we raised venture capital.

Our first customer was my uncle's watch brand, which later became Firebolt. Another was a men's tie brand, and the third was a progressive retail brand that paid us to build custom tablets for their physical stores. These early deals helped us understand real merchant pain points.

In the United States, Shopify succeeded because hiring developers is expensive, so simplicity was highly valued. In India, hiring developers is cheap, so merchants did not care about simplicity. We had to build the websites ourselves, functioning more like a service provider with high gross margins.

When we started, there were no integrations for on-demand logistics, payment gateways, or checkout layers. Payment gateways charged one lakh rupees as a setup fee. We realized we could not just build a store; we had to construct the entire plumbing for open e-commerce.

When we took control of the logistics, our entire mindset shifted because we had skin in the game. Every failed delivery affected our own profit and loss. This forced us to develop address verification intelligence and shipping workflows to reduce failed deliveries.

India is actually ten thousand heterogeneous micro-markets, not one homogeneous block. As our per capita gross domestic product approaches five thousand USD, discretionary spending will inflect. This will drive premiumization and massive demand for regional foods, cosmetics, and specialty goods.

You can scrape subreddits and social media comments using artificial intelligence to analyze exactly what consumers are complaining about. Shiprocket also offers a trends platform that shares data from billions of shipments to show what categories are growing in small cities.

Cash on delivery was introduced by Flipkart to build consumer trust. It mimics the offline counter experience where you only pay when you see the product. It is a terrible supply chain product because it leads to high return rates, but it remains essential for open e-commerce.

We track return history, IP addresses, and shipping patterns. If a buyer has a high return rate, we flag them at checkout. We also check for missing address details, like tower numbers, and prompt the buyer to correct them, reducing failed deliveries.

We built fulfillment hubs to offer cheap two-day delivery, expecting merchants to outsource warehousing. Instead, many merchants refused because they owned their warehouse space and staff, meaning their physical storage cost was effectively zero. We had to adapt to their existing setups.

I would focus on high-margin B2B services using AI. For example, I would build virtual AI voice receptionists for dental clinics or local shops. I would offer the service with zero upfront cost and charge five thousand rupees monthly based on outcomes, scaling up through automated outreach.

Indian merchants are very transactional and cost-sensitive. The only framework that works is proving you can make them twenty thousand rupees if they pay you five thousand. If your product is a vitamin that just makes things look better, rather than a painkiller that saves money, it will not sell.

Unlike software systems that guarantee high uptime, logistics operates in the physical world. Weather, road closures, and delivery issues mean you will inevitably fail on ten percent of shipments. You must educate customers to expect this and manage the fallout.

We successfully trialed drone deliveries in Gurgaon, but the economics do not work for standard last-mile consumer shipping. Currently, drone transport is only viable for high-value, regulated medical cargo where speed is critical and cost is secondary.

Kiranas are exceptionally sharp business operators who understand credit and inventory. Quick commerce is just a Kirana with digital logistics. To survive, Kiranas must be provided with digital toolkits that allow them to manage demand and offer instant deliveries to their local community.