Blinkit vs Zepto vs Swiggy Instamart: What the Filings Actually Say About Unit Economics
By Kartikeya Srivastava | Updated 9 October 2026 | Data: Q1 FY27 (Apr-Jun 2026) filings, Zepto's FY26 UDRHP
Everyone in India's quick commerce race claims they are "close to breakeven." Only one of them is, and the gap between the three is not where most people think it is. This post pulls the latest numbers from Eternal's and Swiggy's Q1 FY27 shareholder letters and Zepto's updated DRHP, puts them on a per-order basis, and tells you what is disclosed, what is not, and what is my own arithmetic.
TL;DR
Blinkit earns about ₹27 contribution per order (5.3% of NOV) and is the only one of the three with positive adjusted EBITDA (₹102 Cr, 0.6% of NOV).
Instamart hit contribution breakeven in May 2026 (-0.2% of GOV for the quarter) but still loses roughly ₹68 per order at the adjusted EBITDA level.
Zepto is unlisted and does not disclose contribution margin. Its UDRHP implies a loss of about ₹79 per order at adjusted EBITDA in FY26.
The Blinkit-Instamart gap is a monetisation gap (about ₹34 per order of gross profit), not a cost gap.Instamart's per-order cost is slightly lower than Blinkit's.
CAC is not disclosed by any of the three. I show the closest proxies below.
Dark-store payback is not disclosed either. Blinkit's numbers let you estimate it (roughly 2 to 2.5 years on contribution at steady state). For Instamart and Zepto, anyone quoting a payback figure is assuming it.
First, why this comparison is harder than it looks
Three problems before any maths:
Different order-value definitions. Eternal reports Blinkit on NOV (net order value). Swiggy reports Instamart on GOV (gross order value) and also discloses NOV. Zepto reports revenue from operations (which includes advertising income). These are not interchangeable.
Zepto is not listed. Its only audited disclosure is the updated DRHP filed in June 2026, and per Moneycontrol's read of it, it leaves out contribution margin, AOV and monthly transacting users, the metrics investors normally want.
Blinkit's revenue is not comparable year on year. It moved from a marketplace to an inventory-led model, so reported adjusted revenue jumped 553% while like-for-like growth was 117%. Use NOV for growth, not revenue.
So: Blinkit and Instamart are compared on Q1 FY27. Zepto is on FY26 (year to March 2026), the latest it has published.
The scoreboard
Metric | Blinkit (Eternal) | Instamart (Swiggy) | Zepto (UDRHP) |
Period | Q1 FY27 | Q1 FY27 | FY26 |
Order value | NOV ₹17,132 Cr (+86% YoY) | GOV ₹7,907 Cr (+40% YoY); NOV ₹5,817 Cr | Revenue ₹22,624 Cr (+104% YoY) |
Orders | 331 M | about 115 M | 640 M (full year) |
Dark stores | 2,443 | 1,171 | 1,139 |
Contribution margin | 5.3% of NOV | -0.2% of GOV (breakeven in May) | Not disclosed |
Adjusted EBITDA | +₹102 Cr (0.6% of NOV) | -₹778 Cr | -₹5,042 Cr (full year) |
Sources: Eternal Q1 FY27 letter, INDmoney's breakdown of it, Swiggy's press release, Inc42 on Zepto's FY26 numbers. Instamart order count is derived from Swiggy's disclosure of about 11.5 crore quarterly orders (BW Retail World).
Contribution margin per order
This is the number that tells you whether an extra order makes money before you pay for head office, tech and ESOPs.
Per order (₹) | Blinkit Q1 FY27 | Instamart Q1 FY27 | Zepto FY26 (my reconstruction) |
Order value | NOV 518 | NOV about 506 (GOV about 688) | Revenue about 354 |
Gross profit / adjusted revenue | about 142 (27.5% of NOV) | 108 (about 21% of NOV) | about 69 (19.6% of revenue) |
Delivery, fulfilment, store and other order costs | about 115 | about 109 | 47.6 delivery + 33.6 dark store = 81 |
Contribution | about +27 | about -1 | about -12 before marketing |
Marketing (A&P) | inside contribution | inside contribution | 21.7 |
Adjusted EBITDA | about +3 | about -68 | -78.75 |
How I got these: Blinkit's gross profit of ₹4,710 Cr and contribution of ₹907 Cr divided by 331 M orders (INDmoney); direct cost of ₹115 per order from Business Standard's coverage. Instamart's ₹108 revenue per order is Swiggy's own number (BW Retail World); the cost line is revenue minus contribution. For Zepto I divided the UDRHP's delivery (₹3,046 Cr), dark-store (₹2,150 Cr) and A&P (₹1,389 Cr) line items by 64 crore orders (Desidime summary of the filing). The ₹78.75 loss per order is as reported by Inc42.
Read this table with care. Blinkit, Swiggy and Zepto classify costs differently, so the lines are directionally comparable, not audited like-for-like. Zepto's contribution is my reconstruction (it ignores packaging, payment fees and other order-level items), so treat it as a ceiling on how good Zepto's contribution could be.
The insight most coverage misses
Look at the cost row. Blinkit spends about ₹115 per order. Instamart spends about ₹109. Instamart is not losing because it is less efficient at delivering a basket. It is losing because each basket earns about ₹34 less gross profit(₹142 vs ₹108).
That matches what Swiggy itself says: it raised revenue per order from ₹83 to ₹108 over five quarters and that is what drove contribution up by ₹28 per order. The lever is mix (higher-margin categories, ads, fees, fewer subsidised baskets), not cheaper riders.
And Blinkit's lead is not secure on the cost side either. Management said contribution was hit by state-level minimum wage increases and the cost of larger new stores, and contribution margin slipped from 5.4% to 5.3% quarter on quarter even as gross margin improved.
CAC: what is disclosed, what is not
Blunt version: none of the three publishes customer acquisition cost. If a deck or a LinkedIn post gives you a CAC for Blinkit or Instamart, it is an estimate. Here is what the filings do let you triangulate.
Blinkit: acquisition spend is buried inside contribution. Eternal defines contribution after delivery, warehousing, wastage, packaging, customer acquisition and payment charges. So a positive 5.3% contribution is after paying to acquire customers. Monthly transacting customers hit 31.8 million (+88% YoY) at the same time, which is the strongest evidence that growth is not being bought at a loss.
Instamart: Swiggy chose to shrink acquisition-led volume. It says it dropped around four million unprofitable users over three quarters, and the platform now has 14 million-plus monthly transacting users. That is a deliberate quality-over-quantity trade, and it explains why GOV grew 40% while Blinkit's NOV grew 86%.
Zepto: the only one with a hard number. FY26 advertising and promotion spend was ₹1,389 Cr, about ₹21.7 per order and roughly 6% of revenue. Its advertising revenue was ₹1,636 Cr, which is more than its own A&P bill (UnlistedZone's analysis). In plain terms, brands paying for shelf visibility on the app are funding more than the cost of acquiring and promoting to users. That is smart, but it also means Zepto's contribution is increasingly dependent on ad income.
Expert read: the useful CAC question is not "how many rupees per install" but "does a cohort pay back its acquisition cost?" Eternal gives the only direct evidence: older customer cohorts spend about 3x what they did three years ago, mostly because they order more often, not because baskets get bigger. Management also expects AOV to stay range-bound. So the whole model rests on frequency, which is a behavioural bet, not a pricing one.
Dark-store economics and payback
Throughput per store
Blinkit | Instamart | Zepto | |
Stores | 2,443 | 1,171 | 1,139 |
NOV per store per day | ₹8.27 lakh | about ₹5.5 lakh (my estimate) | Not disclosed |
Orders per store per day | about 1,600 | about 1,100 | about 2,140 in Q4 FY26 |
Blinkit's number is from Eternal; Instamart's is my estimate from NOV, order count and average stores; Zepto's Q4 figure comes from a secondary read of the IPO filing, so treat it as indicative. Swiggy says network utilisation is around 40%, and over 45% of its stores were contribution-positive in Q1 (up from 30% a quarter earlier).
Blinkit payback, worked through (my estimate)
Eternal discloses enough to estimate this:
Capex per store including warehousing is now guided at ₹2.5 Cr, up from ₹1 Cr.
Net working capital is ₹2,545 Cr, which is about ₹1 Cr per store.
Contribution per store is roughly ₹1.55 Cr a year (₹907 Cr per quarter across about 2,340 average stores, annualised).
Putting ₹2.5 Cr of capex plus about ₹1 Cr of working capital against that contribution gives a payback of roughly 2.3 years on a contribution basis, or about 1.6 years on capex alone. Management's own steady-state case (₹11 lakh NOV per store per day, 4% EBIT on NOV) works out to a 41.7% pre-tax ROCE, which is about the same 2.4-year payback.
Now the caveat that matters. Contribution is before central costs. At today's adjusted EBITDA of ₹102 Cr, the per-store profit is only about ₹17 lakh a year, which on ₹3.5 Cr of capital is a payback of around 20 years. That is not a real payback number, because the network is still ramping (899 stores added in a year) and head-office costs are spread over an immature base. But it tells you exactly what has to happen: throughput must rise about 33% (₹8.27 to ₹11 lakh per store per day) and adjusted EBITDA margin must go from 0.6% to the 6% management guides to.
Instamart and Zepto payback: be sceptical
Instamart: Swiggy does not disclose per-store capex. With contribution at roughly zero, payback on a store is open-ended today. What it does disclose is the path: adjusted EBITDA breakeven needs about ₹60,000 Cr annualised NOV and 5 to 6% contribution (about ₹30 per order), which means quarterly orders going from 11.5 crore to 25 to 30 crore. On a network of around 1,300 stores that is roughly 2,100 to 2,500 orders per store per day, my arithmetic, versus about 1,100 now.
Zepto: no store-level contribution is disclosed. It lost ₹5,042 Cr at adjusted EBITDA on 1,139 stores, and the IPO proceeds are earmarked partly for dark-store expansion and rent. Until it publishes mature-store contribution, treat any Zepto payback claim as marketing.
Market sizing: how big is the prize?
Estimates vary a lot because of definitions, so here is the triangulation.
Source | Estimate |
Redseer | about $13-14 billion GMV in FY26, roughly 17% of online retail GMV (Finnovate summary) |
Redseer, monthly view | about ₹11,000 Cr in a single month in early 2026, up about 100% YoY, on about 7.8 million orders a day (Digital in Asia) |
Datum Intelligence (via Reuters) | 2025 market of about ₹95,500 Cr; Blinkit 46-48%, Instamart 24%, Zepto 22% (StartupFeed) |
Filing-based run-rate | Blinkit Q1 NOV x4 is about ₹68,500 Cr; Instamart GOV x4 is about ₹31,600 Cr; Zepto Q4 FY26 revenue x4 is about ₹30,000 Cr. Together about ₹1.3 lakh Cr, mixing definitions |
Two cross-checks I like. First, Datum's own Q4 FY26 analysis put the exit run-rate of the three at about ₹1.2 lakh Cr, close to my filing-based figure. Second, daily orders from the filings (Blinkit about 3.6 million, Instamart about 1.3 million, Zepto about 2.3 million) add up to roughly 7 million a day for the big three, in the same range as Redseer's all-player estimate.
One number to ignore: Mordor Intelligence's $3.65 billion estimate for 2026 is far below what the listed companies' own filings show, almost certainly because of a narrower definition. Do not cite it next to Redseer without explaining why.
Direction of travel. Datum shows Blinkit at about 51% of the three platforms' combined order value and Instamart's share falling for five straight quarters to about 20%. Growth is shifting to category expansion beyond grocery, and Redseer's forecasts run as high as $30 billion by FY2030, but treat long-range forecasts as scenarios, not facts.
My verdict
Blinkit has proven the model, not yet the returns. Contribution is solid at ₹27 per order. But a 0.6% adjusted EBITDA margin against a 6% target, and capex per store rising 2.5x, mean the next six quarters are about throughput and capital efficiency. Watch whether contribution margin holds near 5% as it adds stores and absorbs wage inflation.
Instamart is fixing the right problem. Revenue per order, not cost, is the gap, and Swiggy has moved it from ₹83 to ₹108. The risk is volume: it needs about 2.5x orders while having shed 4 million users on purpose. If monetisation per order keeps rising without volume stalling, it works. If not, it stays a ₹3,000 Cr-a-year drag.
Zepto is the least visible and the most uncertain. Revenue doubled and loss per order improved from ₹136 to ₹79, but absolute losses still widened to ₹5,905 Cr. SEBI cleared the IPO in May and the UDRHP is on file, but as of early October there is no price band or date. The first listed quarterly disclosure will be the real test.
What to watch next
Q2 FY27 results from Eternal and Swiggy (expected in the coming weeks): Blinkit contribution margin vs 5.3%, NOV per store vs ₹8.27 lakh, Instamart's first full quarter at or above contribution breakeven.
Zepto's price band and whether the roadshow adds contribution margin and AOV disclosure.
Capex per store at Blinkit (₹2.5 Cr) and whether store productivity justifies it.
FAQ
Which quick commerce company is profitable? On adjusted EBITDA, only Blinkit (₹102 Cr in Q1 FY27, 0.6% of NOV). Instamart reached contribution breakeven in May 2026 but lost ₹778 Cr at adjusted EBITDA for the quarter. Zepto lost ₹5,042 Cr in FY26.
What is contribution margin in quick commerce? Revenue minus order-level costs such as delivery, dark-store operations, wastage and payment charges. It excludes head-office costs, so it is always higher than EBITDA margin.
How much does Blinkit make per order? About ₹27 of contribution and about ₹3 of adjusted EBITDA per order in Q1 FY27, on an average net order value of ₹518.
Is Zepto listed? No. It has SEBI's go-ahead and has filed an updated DRHP, but as of early October 2026 it has not announced a price band or IPO dates.
What is a dark store's payback period? Only Blinkit gives enough data to estimate it, and my calculation puts it at roughly 2 to 2.5 years on a contribution basis at steady-state volumes. It is not a company-disclosed figure.
Sources and further reading
Disclaimer: This is analysis for education, not investment advice. Figures marked as estimates or reconstructions are the author's arithmetic from public disclosures and may differ from company definitions. Verify against the original filings before relying on them.