Zeptos Drhp And The Inevitable Cost At The Heart Of Quick Commerce

A couple hundred meters away from the Indiranagar metro station between a juice shop and a car and bike service center is a Blinkit dark store. It’s not very big and it’s not for any of us to browse through. Right outside it, delivery drivers wait on their bikes. Some chat, some scroll, some grab a quick bite before the next delivery and every few minutes some go inside and leave with brown paper packages. In about 10 minutes, someone in a PG or an apartment building will receive their orders of late night ramen or cold meds. Different flavors of cures for the rainy chilly nights right now.

Now, we’ve all heard about and know in theory what dark stores are. There are thousands of them sprinkled across the country, but most of them are concentrated in dense urban residential areas. Most of the time consumers like you and I don’t really give them a second thought. Out of sight, out of mind. But as Zepto’s DRHP filed early this month showed, it’s one of the most core infrastructure for a quick commerce company to scale.

Now, Zepto’s proposed public issue includes a fresh issue of shares worth 8,010 crore rupees and an offer for sale of nearly 11.35 crore shares by existing shareholders. And as Outlook Business reports, a substantial portion of the fresh issue proceeds is directly going to fuel the expansion of Zepto’s dark store network. The company has earmarked more than 1,600 crore rupees to open 1,900 new dark stores across existing and new markets. That’s on top of its existing 1,139 stores, by the way. Now, that’s a staggering total number of 3,039 stores. It’s clear that Zepto is planning on going neck and neck with the current market leader Blinkit. Blinkit’s current CEO Albinder Dhindsa said last October that the company plans on increasing its dark store count to 3,000 by March 2027. And that’s saying something because Blinkit is the sector’s only profitable player with more than 2,200 stores right now. And it is still expanding aggressively because that’s what the model demands.

These small warehouses are an integral piece in the quick commerce puzzle. Scale and success for companies in this space is incredibly dependent on their dark store density. But there’s an uncomfortable truth here that’s rarely discussed. Dark stores are a fixed cost. No matter what happens, whether demands are slow or high or workers are striking, rents have to be paid, warehouse workers have to be salaried, and shelf space has to be filled. The more dark stores that open, the more these fixed costs soar. And there’s still no clear case that exists that proves that the returns are as inevitable as a cost. But there is, in fact, a cautionary tale.

You must remember Dunzo, one of India’s first hyperlocal delivery apps, which had pretty much become a verb. Just Dunzo it was a thing people actually used to say. But despite its popularity, the company shut down its operations in early last year. And analysts believe that part of the reason for its failure was its foray into quick commerce as we know it now. At its peak, it had 120 dark stores operating in 15 cities. But despite its strong brand and clear popularity, it couldn’t afford to keep up with operating and financing this kind of a distributed inventory model.

Even though Dunzo was beaten out by the biggest quick commerce companies of today, they are essentially running the same playbook because they can afford to with their deeper pockets. And it’s no secret that quick commerce as a sector operates with pretty thin margins. A problem that is supposed to be solved when they hit scale. But a large part of the scale bet, which is the dark store model, remains largely unproven. And rising real estate costs, product costs, and a narrowing room to scale is only threatening those thin margins further.

Welcome to Daybreak, a business podcast from The Ken. I’m your host Rishi Verghese, and every day of the week my co-host Nikita Sharma and I will bring you one new story that is worth understanding and worth your time. Today is Wednesday, the 17th of June.

Dark stores aren’t very big. Most of them range from about 600 to 2,000 square feet. And they’re usually densely located close to or in the hearts of residential areas, where the affluent customers who order repeatedly are. Think Juhu or Andheri in Mumbai, Indiranagar or Whitefield in Bangalore, or Greater Kailash or Saket in Delhi. The list goes on. The ideal quick commerce customer is someone who lives in a high-rise or a gated community in any of these locations with enough money to spend on both groceries and a delivery fee, a smartphone, and the unwillingness to go downstairs.

And all these quick commerce orders, along with a 10-minute or close delivery promise they come with, would be impossible without a sound dark store infrastructure. Each of these stores stock about 2,000 to 3,000 SKUs and have delivery radii of 2 to 5 km. They’re all stocked in a way that maximizes efficiency and speed. For instance, when you place an order, it hits the store system instantly. A packer gets the details on a handheld device. The details include what the items are and their locations in the store. The packers then move through the warehouse, scan each item, pack it, and slot it into a numbered pickup point near the entrance. A rider arrives, scans the barcode, picks up the bag, and leaves. They don’t even enter. That’s how the whole thing takes 10 minutes, give or take.

And this hyper-efficient model is what has sprouted this underbelly of a real estate business under the tech layer. Think about it. More than 1,100 physical stores means more than 1,100 individual lease agreements and all the utility and maintenance costs that come with any sort of rented property. But that’s the most visible fixed cost. Because on top of that, there’s the salaries of warehouse managers, the packers who work over multiple shifts, and obviously the delivery partners. There’s also the technology systems like the warehouse management software, inventory tracking, cold storage for all the frozen goods, and of course, even security and surveillance. All these costs have to be paid in full from day one. It doesn’t matter that order density takes months to build.

Now, Zepto’s own claim is that its stores now reach contribution margin positivity in about 6 months, which is down from reportedly 15 to 18 months about 2 years earlier. But contribution margin positive doesn’t equal profitability. Let me quickly explain what that means. It means that the money coming in from orders at that store covers the direct cost of fulfilling those orders with some left over. Now, to be fair, being contribution margin positive is in fact real progress. It means that the store isn’t losing money on every single delivery anymore and all the direct costs are being covered. But, the overhead costs keep running regardless. And the argument for how scale would eventually minimize those costs has a lot to prove. More on this in the next segment.

Here’s the main argument that justifies the significant cost of a dark store or hundreds of it. With an eventual increase in order density, dark stores can process more orders without a proportional increase in operating expenses through improved labor productivity and better utilization of infrastructure. So, the fixed costs like rent, staff, etc. get spread across orders. And if you factor in the shorter delivery radii that comes with more dark stores, that also means lower per order last mile cost.

Zepto, for its worth, has already started to work towards reducing its per order cost. Inc42 reported that while a majority of quick commerce startups primarily focus on geographic expansion, Zepto is working on density instead. It aims to have clusters of dark stores within its existing markets. And the startup believes that this strategy enables shorter delivery distances, higher order throughput, and lower fulfillment costs. It’s also staying in the urban areas where demand is already strong. Instead of spreading itself thin. Also, the company has started bundling orders in the same neighborhood through a single delivery executive. This does improve the efficiency of the process and lowers the per delivery cost.

And both moves have been showing results. Zepto’s orders processed per day per store or OPD increased by more than 50%. It jumped from close to 1,500 to over 2,000 between March 2025 and March 2026. The order bundling system has also resulted in a drop in Zepto’s supply chain variable cost per order, which has started to inch down from 63 to 61 rupees over the past 2 years. They’re small numbers, but at least they’re moving in the right direction.

But there’s still a catch. You know, of course this model works perfectly in the densest parts of Delhi NCR or Mumbai. Stores get upwards of 2,000 orders a day, and several of these orders would be for high-margin items. Think premium cold cuts, artisanal sourdough, cheeses, coffee beans, and your oyster sauces or packaged kimchi. Orders that include these kind of items are far more profitable for a quick commerce platform than an order that’s just milk and bread and eggs. Basically, urban affluent customers are quite important to this model, which is why the items I just listed cater not just to the people who shop at Kiranas, but also at the high-end grocery stores like Nature’s Basket, for example. The hope for these platforms is that well-off users like this will continue to purchase high-margin products like this and push up the overall value of each order.

But here, the challenge begins as well. One is the fact that India continues to be a highly price-sensitive market. That’s pretty well documented. For example, a report from the India Brand Equity Foundation noted that the price-sensitive nature of India’s market poses sustainability challenges for quick commerce players in the long term. Which means if delivery costs go up and discounts start disappearing, most likely so will the customers. And since Zepto functions on an inventory-led model, it has to bear the losses for all the high-margin items that don’t sell on its platform. This also reduces the scope for scale, which is one of the biggest projected factors for future profitability. You’ve heard it. Once we reach scale, we’ll become profitable.

You see, there’s only so many urban affluent customers to tap in on, and only so much real estate in cramped cities for aggressive expansions. And expanding out of these geographies comes with very real risks. In tier two cities, where transport connectivity isn’t as good, where demand isn’t as established, and consumers are likely to be price sensitive, the dark store push would have to go hand in hand with customer acquisition costs. And that’s already showing signs of slow growth. Reportedly, despite spending more than 1,300 crore rupees on advertising and promotional activities in FY26, Zepto actually recorded a decline in its user base from December 2025 to March 2026 by more than a million users.

Then, there’s the squeeze on the margins that’s already happening. Let’s take real estate prices. In the very cities where all these quick commerce companies are fighting for density, urban commercial rents have only been rising. The Print reported this February that quick commerce companies are driving up commercial real estate prices by paying higher than the asking price in their desperation to lock down appropriate locations. Here, companies with deeper pockets have the obvious advantage. Vacancy rates in cities like Mumbai and Bangalore are always low, about 3 to 5%, which means asking prices are already quite steep. Add to that the premium companies are paying to beat out competition, and it’s already a sizable investment. In fact, UBS, a financial services company, estimates the average dark store to cost anywhere between 45 to 60 lakh rupees in CAPEX. These costs are only going to increase as time goes on, even as newer competitors like Flipkart and Amazon join the race with more aggression.

And rents aren’t the only rising costs. There’s also general inflation and the price hikes many FMCG goods have seen since the US-Israel war on Iran began. Several companies took on hikes of about 3 to 10% to protect their own margins. Which means quick commerce platforms are either absorbing that cost or passing it on to consumers. Absorbing the cost obviously erodes margins because the platform is making less from each order. Meanwhile, passing it on ties into the price sensitivity we spoke about earlier. Customers will either pick cheaper options or buy less. And that would reduce the overall order value as well.

And what’s not very encouraging about the whole thing as well is how the only profitable player in the sector is doing. Now, Blinkit has been profitable since 2024. But an Entracker report more recently noted that the company’s profitability remains quite thin. Despite operating over 2,000 dark stores with 17 million square feet of space and deploying over 4 lakh delivery partners, the business reported an adjusted EBITDA of just 37 crore rupees. And that actually amounts to a margin of around just 0.3% of net order value.

So, what exactly is Zepto working towards? Now, its own DRHP meanwhile shows that while the company is growing rapidly in revenue, it’s also losing more than before. The DRHP even states that the company actually expects to keep incurring losses in the near future. The company’s operating revenue rose to nearly 23,000 crore rupees in FY26, up from a little more than 11,000 crore rupees in FY25. But at the same time, net loss increased to almost 6,000 crore rupees in FY26 compared to about 5,000 crore rupees in FY25.

Considering the steep competition Zepto is going to face and quite literally the limited room to expand in urban areas, it’s clear that the entire business model stands on shaky ground. As long as the model depends on acquiring fixed cost physical infrastructure, the bet is that the business will make revenue long enough until the infrastructure starts to make back its cost. But that bet is not a promise.

Daybreak is produced from the newsroom of The Ken, India’s first subscriber-focused business news platform. What you’re listening to is just a small sample of our subscriber-only offerings. A full subscription offers daily long-form feature stories, newsletters, and a whole bunch of premium podcasts. To subscribe, head to theken.com and click on the red subscribe button on the top of the Ken website. Today’s episode was hosted and produced by my colleague Rachel Verghese and edited by Rajiv Sian.