Widely read:
Ujjal Das/AI-Generated Image
Summary: Private investors valued Zepto at one number. Public market investors had something very different in mind. The gap says less about quick commerce and more about how businesses are really priced when buyers are finally free to say no.
India's largest mutual funds were asked what Zepto is worth. They said about a third of what its last private investors had paid. The listing has been deferred, and retail investors were never asked to fill the gap.
That is the whole lesson, and it matters more than any view on quick commerce. You are allowed to refuse a price. Everyone else in the chain has a reason not to.
The news
Zepto filed its updated draft prospectus on June 8, 2026, seeking Rs 8,010 crore of fresh capital plus an offer for sale of 11.35 crore shares by six early investors. It wanted a July listing. By late July, domestic institutions were indicating $2.5 billion to $3 billion, against the $7 billion at which CalPERS led a round in October 2025. Zepto has stepped back and is talking to existing backers about a little over Rs 1,000 crore instead. Its papers stay valid until around August 21.
The $7 billion was never a price
It was a number agreed among a small group of people who all had an interest in the number being large. The founders, the existing investors, the incoming investor that would carry the stake at that mark in its own books. Nobody in that room argued for less.
An IPO price is different. It is the first time a wide set of buyers who owe the company nothing is asked what the business is worth. The answer came back at roughly a third.
That is not the market being unfair. That is the market working.
The domestic mutual funds have done retail investors a service here. The old complaint was that they signed off on whatever price the bankers put in front of them. This time one of them reportedly did not bother to make a counter-offer.
The gap between what the last private round said and what the public market says is the most useful thing to come out of this episode, and it did not come from a prospectus.
Twice in five months
PhonePe deferred its offer in March 2026, after its bankers were reported to have recommended $9 billion to $10.5 billion against a whisper number of $15 billion. PhonePe attributed the pause to market conditions rather than pricing, and that distinction deserves respect.
Set the stated reasons aside and look at what happened. Twice this year, a company arrived with a private mark, institutions indicated a number far below it, and the company chose to wait. For anyone raising more than about half a billion dollars in India, the bar has moved. A private valuation is now treated as what it is, the seller's opening position.
What the funds were reading
Zepto's revenue from operations more than doubled in FY26 to Rs 22,624 crore from Rs 11,110 crore. The loss grew to Rs 5,905 crore from Rs 4,700 crore. Operations consumed Rs 3,462 crore of cash. Cash on the books fell to Rs 5,681 crore in March 2026 from Rs 7,441 crore a year earlier.
The operating record is improving. Zepto ran 1,139 dark stores across 66 cities in March 2026, against 337 stores in 11 cities two years earlier. The adjusted loss per order in the March quarter narrowed to Rs 59.4 from Rs 142.68 a year before.
Both things are true at once. The business is getting better per order and losing more in total, because it adds orders faster than it fixes the economics of each one. Buying at $7 billion means believing the second curve crosses the first before the cash runs out. Institutions decided that belief was worth $3 billion.
The structure of the offer says the same thing. Zepto is listing under Regulation 6(2), the route for companies that fail the eligibility tests in Regulation 6(1), usually the requirement of Rs 15 crore of average operating profit over three years. Under 6(2), at least 75 per cent of the net offer must go to institutions and retail gets no more than 10 per cent. On the smaller Rs 5,000 crore issue discussed since, retail would have been about Rs 500 crore. Retail was never setting this price. The funds were. Remember that the next time a loss-making issue is sold as a chance for the small investor.
A deferral buys time, not economics
Zepto lost close to Rs 5,900 crore last year and will keep burning cash whether or not it lists. The wager is that another year of better unit economics fetches a better number. It might.
But the company must fund itself meanwhile, and the arithmetic is unforgiving. That Rs 3,462 crore of cash used in operations works out to roughly Rs 865 crore a quarter. The round being discussed is a little over Rs 1,000 crore. On last year's run rate, it buys one quarter.
Raising privately at about $4 billion after public investors have indicated $2.5 billion to $3 billion does not settle the argument. It postpones it.
There is a harder problem underneath. The things that would fix the unit economics are the things that slow the growth: fewer new stores, less discounting, higher delivery charges, tighter marketing. Each improves the loss per order. Each also erodes the growth rate the valuation was built on. The next two or three quarters will show whether that trade can be made without breaking the story.
Three rules that survive every cycle
An IPO is a sale, and the seller picks the moment. No company lists when its own managers think the shares are cheap. The burden of proof sits with the buyer.
Judge the price, not the structure. Much Indian commentary blames the offer for sale, as though existing shareholders selling is the disease. Some of the best listings of the last three years were largely offers for sale, and some of the worst were pure fresh issues. What separates them is the price you pay against the earnings the business can produce.
There is no urgency. A good business will be for sale on any Tuesday for the next 20 years. The fear of missing a listing gain is the most expensive emotion in the primary market.
The market you are buying into
We track every public issue in India. Between December 4, 2024 and July 31, 2026, 563 issues came to market, 396 of them on the SME platforms and 167 on the mainboard. In calendar 2025, 107 mainboard issues raised Rs 1,83,258 crore at a median size of Rs 745 crore, alongside 271 SME issues raising Rs 12,691 crore at a median of Rs 39 crore. In 2026 so far, 42 mainboard issues have raised Rs 66,868 crore, with 108 more SME issues.
An Indian investor was offered a new company on roughly three of every four trading days. Nobody can evaluate businesses at that rate, including us. The only defence is to decline most of them without guilt.
From the same data: offers for sale were 45 per cent of all shares sold in mainboard IPOs in 2025, and 17 of the 107 issues were entirely offers for sale, with no money reaching the company. In 2026, the figure is 47 per cent. That does not make those issues bad. It tells you what the primary market mostly is right now, a way for early shareholders to sell to new ones.
Source: Value Research primary market database, issues opening December 4, 2024 to July 31, 2026.
Four checks before you apply for anything
Find the eligibility route on page one. Regulation 6(2) means the company failed the tests a normal listing requires. A fact, not a verdict, but know it first.
Find the cost of acquisition, and read both versions. The cover page carries each selling shareholder's average cost. Nexus Ventures VI's is Rs 3.91 a share, Contrary's Rs 3.98, Nexus Ventures VII's Rs 23.65. Those are lifetime averages assuming full conversion of preference shares. The Basis for Offer Price section carries a second figure, the weighted average cost of acquisition over the trailing eighteen months, which shows what money was paid recently rather than in 2021. Read both, and ask why they differ.
Check whether the business has produced cash from operations. Not adjusted profit, not store-level contribution. Cash. It is in the restated cash flow statement.
Watch the institutional book. Under Regulation 6(2), the issue fails outright unless institutions take 75 per cent, and the money goes back. They do not set the price band. They decide whether it holds. Anchor allocations appear on the exchange websites the day before the issue opens, and subscription runs live by category during it.
What to carry away
The funds that refused this price hold your SIP money. They were doing your job, with your money, as your agents, and this month they earned their fee.
They will not always be there. Most of the issues in our data were on the SME platforms, where no large institutional book stands between the seller's price and yours. In a market offering a new company three days out of four, the skill worth building is not picking the right IPO. It is saying no to most of them.
Also read: Two of every three IPO rupees fund someone else's exit
Ask Value Research 
New episodes every Friday 
New Process Alert 
Seems like you've uploaded a wrong QR code. Please upload the QR code downloaded from MF Central platform.
