Fractal Analytics: Is India Finally Ready For True Technology IPOs; Or Are We Seeing Overvaluation Based On Hype?

On February 16, 2026, Dalal Street was supposed to celebrate a watershed moment. Fractal Analytics — a 25-year-old enterprise AI firm, a company that counts Citi, Nestlé, Microsoft, and Nvidia among its clients — was finally going public. Newspapers called it India’s AI coming-of-age story. Bankers called it a bellwether for the country’s deep tech ambitions. The government, busy hosting an AI Impact Summit in New Delhi that very week, couldn’t have scripted a better backdrop. But, what actually happened was something far more reverse!

Fractal Analytics shares debuted at ₹876 on the NSE — a 2.67% discount to its issue price of ₹900 — and quickly deteriorated, hitting an intraday low of ₹852 as heavy selling pressure set in. Within the first few hours of trading, over ₹700 crore in notional investor wealth had been erased. This was not a market overreaction. It was the market doing exactly what it is supposed to do: pricing reality.

As of March 3, 2026, the stock is trading at ₹751 and market cap is ₹12,923 crore per Screener and 5paisa live data. For investors who applied at ₹900, that is a clean, calculable loss. And for anyone who has watched this movie before, and India has watched it many, many times, the plot was not difficult to predict.

The Anatomy of a Decorated Exit

Let’s be precise about what Fractal Analytics’ IPO actually was, because the language of “India’s first pure-play AI IPO” has done considerable work obscuring some very basic arithmetic. The total issue size was ₹2,833.90 crore. Of this, the fresh issue — money that actually goes to the company — was ₹1,025.58 crore. The Offer for Sale component, money that goes to existing shareholders exiting their positions, was ₹1,808.32 crore. That means 63.8% of the IPO proceeds went straight to selling shareholders, not to build products, not to hire engineers, not to expand into new markets, but to exit.

The selling shareholders include Venkateswara Remala, Satya Kumari Remala Rao, GLM Family Trust, Quinag Bidco, and TPG Fett Holdings Pte Ltd. These are not small names — Quinag Bidco and TPG Fett are institutional investors who have held stakes in Fractal for years. They came to the IPO to monetise. Retail investors came to participate in India’s AI story. These are not the same transaction, even if they occur on the same form.

This is not a new trick. Data from Prime Database shows that the share of OFS in Indian IPOs has been nearly 73% of total proceeds in recent years, up from a historical balance between fresh issues and secondary sales. Early backers of companies like Paytm acquired shares at ₹15.40 while Nykaa’s early investors bought in at ₹76.65 — and both groups sold at IPO prices of ₹2,150 and ₹1,125 respectively. The math works extraordinarily well — for them. Retail investors standing at the other end of that transaction have consistently been left holding a depreciating asset.

Is The Valuation Be Justified?

Fractal’s pricing was not conservative, perhaps, it was not even defensible by traditional metrics.

At the IPO price of ₹900, the implied Price-to-Earnings ratio (Pre-IPO P/E) on FY25 profit was approximately 67.37x, already placing Fractal at a significant premium to the broader market, with Nifty 50 trading at a P/E of roughly 22.35 as of early February 2026. Even at the listing price of ₹876, the implied P/E on FY25 profit remained at 65.6x. This is not the valuation of a company that is on stable financial footing. This is the valuation of a company being priced for a future that has not yet arrived!

Consider the actual financial picture beneath the AI narrative. Total income for the six months ended September 30, 2025 was ₹1,594.30 crore, with a profit after tax of ₹70.90 crore. The EBITDA margin stood at 11.90%, and the return on net worth was a modest 3.6%. These are not the numbers of a high-growth, high-margin technology platform. These are the numbers of a consulting-adjacent services firm with AI branding applied on top.

The revenue quality is similarly worth examining. The top 10 clients accounted for 54.2% of Fractal.ai segment revenue in the six months ended September 2025, and the US accounted for 64.9% of total revenue in the same period. This is not a diversified, platform-driven technology business. This is a B2B services company with deep client concentration, meaningful geographic risk, and an overwhelming dependence on continued US enterprise spending; which, under the current global economic uncertainty, is far from guaranteed.

Employee benefit costs were 72.2% of revenue for the six months ended September 2025. In other words, for every ₹100 Fractal earns, ₹72 goes to paying people. This is the structural cost profile of a manpower-intensive services business, (perhaps) not a scalable AI platform with software-like margin expansion. The IPO was priced as though Fractal is the latter. The financials confirm it is the former.

A Pattern India Refuses to Acknowledge

The tragedy of Fractal Analytics is not that it had a bad listing. It’s that India has seen this exact story play out — with different names, different sectors, and identical outcomes — so many times that the pattern should be embarrassing.

Paytm launched India’s largest-ever IPO at ₹2,150 per share in November 2021. On its first day of trading, shares sank to ₹1,560 — a 27% crash that wiped out billions in market value within hours. The company’s founder declared that the listing was “no indicator of the value of our company.” It was, in fact, an extremely accurate indicator.

Five of India’s most hyped technology IPOs — Paytm, Zomato, Nykaa, Delhivery, and PolicyBazaar — collectively shed more than $18 billion in value after listing, as investors who piled in at elevated IPO prices were left with catastrophic losses. Portfolio managers who watched the collapse were blunt in their diagnosis: “Valuations of these companies were not supported by fundamentals and the balance sheets, and their cash burn was high.”

If you had invested ₹10,000 in the 2021 tech IPO rush on the first day of trading, your investment would be worth ₹3,302 in Zomato, ₹3,597 in PolicyBazaar, ₹4,434 in Paytm, ₹4,470 in CarTrade, and ₹6,137 in Nykaa — within just twelve months. An investor who spread their money equally across these five “growth stories” would have lost, on average, more than half their capital in a single year. Not from a scam. Not from fraud. From IPOs that were priced with intention at valuations the market was never going to sustain.

Retail investors often believe their money is going into well-researched, blue-chip companies through IPO allocations; however, in reality, a significant portion is being quietly funnelled into risky, unproven startups at peak valuations. This is not financial education failure, but is information asymmetry by design!

Would It Be Wrong If We Say Fractal’s IPO Shrunk Before It Could Embarrass Itself Further?

There is one detail about Fractal’s IPO that deserves far more attention than it has received.

In early February 2026, the IPO size was cut by more than 40% — from an originally planned ₹4,900 crore down to ₹2,834 crore — on the advice of the company’s own bankers. Let that sit for a moment. The very people whose financial interests are tied to maximising the IPO size told the company to take 40% less money from the market. That is not a routine adjustment. That is a signal from professionals with real-time market intelligence, that even at a reduced size, this was going to be a difficult sale.

Even after the cut, Fractal’s public market valuation was lower than its private valuation of $2.4 billion from a secondary sale in July 2025. This is the modern definition of a down-round IPO; a company coming to the public market at a valuation below what sophisticated private investors paid months earlier. The retail investor, armed with a prospectus dense with AI terminology and a marquee client list, has no obvious mechanism to understand that they are paying less than institutional investors recently did, and still overpaying relative to what the earnings can support.

The Deeper Question: What Is a Technology IPO in India?

India has spent five years celebrating the democratisation of its capital markets. Retail investor participation has grown dramatically. DEMAT accounts have multiplied. IPO subscriptions routinely cross 20x and 30x. The country has built a genuinely accessible, genuinely participatory public market.

And then it has systematically used that market to transfer wealth from retail investors to venture capitalists.

Experts have raised an uncomfortable question that the ecosystem has largely refused to answer: are some of these companies being built not to last, but just to list? The evidence suggests this is not a fringe concern. When OFS routinely exceeds fresh issue components, when anchor lock-in expiries are circled on every analyst’s calendar as selling opportunities, when IPO pricing routinely exceeds any defensible earnings multiple, the IPO is not a growth event. It is a distribution mechanism.

Fractal’s revenue grew from ₹1,985 crore in FY23 to ₹2,765 crore in FY25, and the company swung from a net loss of ₹54.70 crore in FY24 to a net profit of ₹220.60 crore in FY25. This is real progress, and Fractal is not a fraudulent company. But progress and IPO-worthy valuation are different things. A single year of profitability after years of losses, priced at 67x earnings in a market where the benchmark index trades at 22x, is not an investment case; perhaps it is a leap of faith being sold as a certainty.

Anchor lock-in expiry dates are March 13, 2026 and May 12, 2026, with many IPOs historically seeing 5–15% selling pressure around such windows. Retail investors who bought at ₹900 should be watching those dates, not the AI headlines.

At The End, India Deserves Better Than This!

India is not ready for true technology IPOs, perhaps; not because its companies aren’t talented enough, but because its market infrastructure keeps rewarding the wrong behaviours.

A true technology IPO is priced to reflect sustainable earnings power, not peak hype. It has a fresh issue that meaningfully exceeds OFS, because the company needs capital to grow, not because investors need an exit. It comes to market when the business model is proven, not when the fund life is expiring.

What India has instead is a pipeline of VC-funded companies racing to list before the music stops, priced by bankers incentivised to maximise deal size, sold to retail investors who have been told that participating in the stock market is the same as participating in India’s economic growth.

Fractal Analytics may yet prove the market wrong. Its client list is real. Its AI capabilities have a genuine foundation. With a net revenue retention rate of 121% in FY25 and a position in the $310 billion global data-AI services market, there is a long-term case to be made. But that case was not worth ₹900 per share on February 16, 2026. The market said so clearly, immediately, and without apology.

All financial data cited in this article is sourced from Fractal Analytics’ IPO filings, BSE/NSE exchange disclosures, and publicly available market data. This article represents the editorial opinion and does not constitute investment advice.