Quick Commerce Unit Economics: The Numbers They Buried in Plain Sight

Here is the quick commerce unit economics nobody puts on a pitch deck. In Q4 FY26, Blinkit, the “profitable” one, earned about Rs 1.35 of adjusted EBITDA per order. Instamart lost roughly Rs 76 per order. Zepto, still private, burned a Rs 5,905 crore net loss across FY26. So the model that promised your groceries in ten minutes makes almost nothing when it wins, and haemorrhages cash when it does not. The winners and losers run the same math. The difference is who admits it.

Everyone keeps calling quick commerce a rocket ship. The filings call it a treadmill.

The story you have been sold is simple. Zepto, Blinkit and Instamart cracked ten-minute delivery, Indians got addicted, and now the money prints itself.

The numbers say something less flattering. When this business wins, it wins by rupees per order. When it loses, it loses by thousands of crores a year. There is no comfortable middle. That is the whole story, and the companies would rather you looked at the growth charts instead.

Let us look at the boring part they hope you skip.

What does “profitable” actually mean at Blinkit?

Blinkit is the poster child. Its parent, Eternal Ltd (the company formerly known as Zomato), reported that Blinkit turned adjusted EBITDA positive in Q4 FY26, posting Rs 37 crore. That is a real milestone. It is also a rounding error.

Blinkit’s net order value that quarter was Rs 14,386 crore. So Rs 37 crore of adjusted EBITDA is about 0.3% of the value flowing through the platform. For every Rs 100 of groceries Blinkit moved, it kept roughly 30 paise.

Now do it per order. Blinkit’s net average order value was Rs 525. Divide Rs 14,386 crore of order value by a Rs 525 basket and you get roughly 274 million orders in the quarter, on our own arithmetic from Blinkit’s disclosed figures. Split Rs 37 crore across 274 million orders and Blinkit earned about Rs 1.35 per order.

Read that again. The most celebrated quick commerce business in India makes less than the cost of a single Parle-G packet on each delivery.

This is the same structural squeeze The Brand Crush flagged when we asked whether all three players could win at once. The unit economics answer is now on paper.

---

---

Why is Instamart still bleeding on every order?

Swiggy is publicly listed too, so Instamart’s numbers are out in the open.

In Q4 FY26, Instamart’s gross order value grew a punchy 68.8% year on year to Rs 7,881 crore. Growth is not the problem. Growth was never the problem. The problem is the margin underneath it.

Instamart’s adjusted EBITDA loss was Rs 858 crore for the quarter, at a margin of negative 10.9% of gross order value. Per order, the picture is brutal. Instamart’s average order value was Rs 700. Rs 7,881 crore of order value at a Rs 700 basket is roughly 112 million orders, again our own calculation from Swiggy’s disclosed figures. Spread an Rs 858 crore loss across those orders and Instamart lost about Rs 76 on every single delivery.

Swiggy’s own answer is scale. Management says Instamart can eventually become a Rs 1 lakh crore net-order-value platform at 4% to 5% EBITDA margins. Maybe. That is a bet on a future basket, not a description of today’s one.

---

So where does that leave Zepto?

Zepto is the one that does not have to show its per-order homework, because it is still private.

What it cannot hide are its IPO filings. Zepto’s operating revenue roughly doubled to Rs 22,624 crore in FY26, up from Rs 11,110 crore in FY25. Impressive. Its net loss for FY26 was Rs 5,905 crore.

Let that sit next to the revenue. Zepto is spending nearly six thousand crore of losses to buy that growth. Delivery and handling alone cost Rs 3,046 crore across the year. It runs about 1,139 dark stores as of March 2026, and it does not publish an adjusted per-order EBITDA the way the listed players do. You can see why. We broke down the same gap between headline revenue and real margin in our look at Zepto’s valuation versus its average order.

---

Can dark-store density fix the math?

This is the industry’s favourite defence, and it is not nothing.

A new dark store opens empty and loses money. It takes months to build the local order density that covers its rent and its pickers. So a fast-expanding network drags down the average, because the young stores bleed while the old ones earn.

Blinkit is the proof it can turn. One year earlier, in Q4 FY25, Blinkit was posting a Rs 178 crore adjusted EBITDA loss. Twelve months and 95% order-value growth later, it scraped to a Rs 37 crore profit. Older stores maturing is a big part of that swing. Eternal is now targeting one billion dollars of adjusted EBITDA from its quick commerce and going-out businesses by FY29.

Here is the catch. Density fixes the store, not the model. Even mature, Blinkit’s blended margin is 0.3%. The bet is that a bigger basket, more ad income, and denser routes stack into a real margin later. Nobody has proven that ceiling yet. They have only proven you can stop losing money, barely.

---

Is “adjusted EBITDA” hiding the real number?

Yes, and you should treat that phrase with suspicion every time you see it.

Adjusted EBITDA strips out employee stock costs, depreciation on all those dark stores, and interest. In an asset-light software business, that is a fair proxy for cash. In quick commerce, you are building physical warehouses across hundreds of cities. The depreciation and the capex are the business, not a footnote.

Call something “profitable” while you are still pouring capital into the machine that makes the profit, and you are using the most generous word available. None of this is fraud. Every number here comes from the companies’ own filings. The trick is not in the numbers. The trick is in which number gets the headline. This is the same playbook we saw in the super app fantasy: grand framing, quiet math.

---

The verdict

Quick commerce in India is not a money machine. It is a market-share machine that occasionally, at its absolute best, breaks even by a rupee or two per order.

Blinkit earns about Rs 1.35 an order and gets called the winner. Instamart loses about Rs 76 an order and calls it progress. Zepto loses Rs 5,905 crore a year and calls it scale.

Someone will eventually make real money here. It will be whoever gets the basket big enough, the stores dense enough, and the ad income high enough to stop depending on your groceries for margin. Until then, the ten-minute promise is being subsidised by investors, not earned from you. Enjoy the convenience. Just know it is running at a loss so precise you could measure it in biscuits.

---

FAQ

Is Blinkit actually profitable?

Only on an adjusted basis, and barely. Blinkit posted Rs 37 crore of adjusted EBITDA in Q4 FY26, worth about 0.3% of its order value. Adjusted EBITDA excludes depreciation on dark stores and stock-based pay, so it is not the same as net profit.

Which quick commerce company loses the most money?

By absolute annual loss, Zepto: a Rs 5,905 crore net loss in FY26. By per-order loss among the listed players, Instamart lost roughly Rs 76 an order in Q4 FY26 on our arithmetic from its disclosed figures.

What is a good average order value for quick commerce?

Higher is better, because the delivery cost barely changes with basket size. In Q4 FY26, Instamart’s average order value was Rs 700 and Blinkit’s net average order value was Rs 525. Both companies are pushing baskets up to fix the margin math.

Why does quick commerce lose money if it is so popular?

Popularity is not the problem. Each order must cover a dark store, inventory, a picker and a rider for a small basket delivered in ten minutes. That fixed cost per drop is high, so margins stay thin or negative until baskets and store density rise.

Can quick commerce ever be truly profitable in India?

Possibly, at scale. Swiggy targets 4% to 5% EBITDA margins if Instamart reaches Rs 1 lakh crore of net order value, and Eternal targets one billion dollars of adjusted EBITDA by FY29. Both are future promises, not current results.

The polish says “profitable.” The filings say Rs 1.35 an order. The Brand Crush reads the footnotes so you do not have to. No brand pays us. Nobody reviews us before it goes live. Subscribe for the real numbers behind the campaigns.

The Brand Crush publishes independent brand analysis and fair comment based on public filings and reporting. Financial figures are drawn from company disclosures and named media sources cited above. Per-order figures are our own arithmetic from disclosed order value, average order value and adjusted EBITDA, and are labelled as such. Nothing here is investment advice.