India's leading pure-play enterprise data, analytics, and AI company, Fractal Analytics, is set to tap the capital markets through an initial public offering (IPO) that opens on 9 February. Fractal plans to raise ₹2,834 crore, at a price band of ₹857-900 per share. The issue comprises a fresh issue of ₹1,024 crore and an offer-for-sale of ₹1,810 crore by existing investors, including Quinag Bidco (Apax Partners), TPG Fett Holdings, and the GLM Family Trust.
At the upper end of the price band, Fractal's implied market capitalization stands at ₹15,480 crore. This IPO offers investors exposure to an integrated AI provider serving Fortune 500 clients, a unique proposition given that there are no listed peers in India with a comparable business model.
Capital allocation: Investing in Alpha and infrastructure
Of the net proceeds, Fractal plans to deploy ₹355 crore towards investment in sales, marketing, and research and development (R&D) for its Fractal Alpha segment, ₹265 crore to repay borrowings of its subsidiary, Fractal USA, and ₹121 crore to set up new office premises in India.
Additionally, ₹57 crore will be used to purchase laptops, with the balance to be utilized for inorganic acquisitions and general corporate purposes. This allocation underscores a strategy focused on product innovation and operational expansion to deepen its AI capabilities. The larger question, however, is how Fractal stacks up operationally and financially.
The Fractal.ai and Alpha model
Fractal organizes its operations into two primary segments: Fractal.ai (core business) and Fractal Alpha (incubation and new ventures) to serve diverse client needs. Fractal.ai provides AI services and products designed to build customized solutions for clients. Many of its AI products are hosted on Cogentiq, the company's flagship agentic AI platform. This segment is the company's primary cash-flow generator.
Cogentiq provides prebuilt agents, tools, and connectors to help enterprises accelerate the development and enhancement of AI applications. The segment generated ₹2,704 crore, accounting for nearly 98% of FY25 revenue of ₹2,765 crore.
Fractal Alpha, on the other hand, is an innovation engine that focuses on incubating new ideas and scaling independent AI businesses. Within this segment, successful products or ideas with high market potential are developed as standalone businesses under separate management.
While smaller in absolute terms, the Fractal Alpha segment is growing rapidly. Revenue for this segment grew by 76% year-on-year to ₹64 crore in FY25, and by 64.0% in H1FY26. However, in investment mode right now, this segment is incurring losses.
R&D edge
This is why Fractal has consistently invested 5-7% of its revenue in R&D to enhance its AI solutions and products. The technical validation of this innovation engine is already visible. Fractal has been handpicked for the IndiaAI Mission to build a sovereign reasoning model for the nation.
This initiative serves as a flagship for the Fractal Alpha segment, demonstrating its ability to move beyond standard enterprise services into foundational AI development that could eventually be delivered globally. Still, it believes that artificial general intelligence could surpass its own AI capabilities, posing a threat to its business model and competitive position.
Defending the AI moat
Fractal also relies on trademarks and patents to protect its technical know-how and intellectual property rights. As of 19 January 2026, it has 29 registered and 38 pending patents, along with 376 registered and 104 pending trademarks.
Fractal generates revenue through multiple pricing models depending on the engagement. This includes time-and-materials (based on time and personnel resources deployed), fixed-price (a fixed fee for projects or retainerships), subscription/licensing, and output-based models. Given the AI landscape, R&D is essential for survival in the industry.
The high-stakes 'must-win' client model
Fractal targets large global enterprises (Fortune 500-equivalent) with revenues over $10 billion, market caps over $20 billion, or end-customer bases over 30 million. The company refers to such clients as Must-Win Clients (MWCs). As of September 2025, it served 122 such clients, including a majority of the “Magnificent Seven” technology giants, such as Google, Microsoft, Amazon, and other global enterprises.
Concentration risk
The top 10 clients accounted for 54% of revenue in the Fractal.ai segment in H1FY26. However, the largest client accounted for 8% of revenue, indicating moderate dependence on individual customers. Clients are sticky, with relationships with the top 10 clients averaging more than eight years. This longevity, however, limits the company's pricing power.
Most clients (52) generate over $1 million in annual revenue, 17 over $5 million, 7 over $10 million, and 4 over $20 million as of H1FY26. This strategy has proven effective, evidenced by a 114% net revenue retention rate, meaning existing clients are spending 14% more year-on-year. However, the loss of any key MWC relationship could severely impact financials.
For example, in 2022, one of its top 10 clients in the banking, financial services, and insurance (BFSI) industry re-evaluated its essential and non-essential spending as part of an internal restructuring. Sonata Software, a Microsoft software reseller, provides another example: its Q2FY26 revenue declined by 30% after Microsoft began selling licenses directly to clients.
The global footprint
MWCs accounted for 79.6% of Fractal.ai segment revenue in H1FY26. This also means that the company generates the vast majority (92.4%) of its revenue from outside India. This exposes Fractal to geopolitical and regulatory risks, similar to those recently faced by IT companies with a significant presence in the US. Like them, Fractal derives around 65% of revenue from the Americas, followed by Europe (21%), India (8%), and Others (6.0%) in H1FY26.
Diversified revenue-mix
Fractal's revenue is diversified across four key verticals. In the revenue mix, consumer packaged goods and retail (CPGR) accounted for 38% of revenue, followed by technology, media and telecom (TMT) (27%), healthcare and life sciences (17%), BFSI (12%), and others (energy, travel, and industrial). A cyclical downturn in any of these verticals could weigh on growth.
As an AI company handling sensitive client data, Fractal faces risks from cyberattacks, data breaches, and evolving privacy laws. The company experienced a ransomware attack in June 2020, which resulted in the termination of a client contract and the loss of a potential client. The pace of change in the AI industry further compounds this risk. Failure to adapt to new technologies, including AGI, or shifting client preferences could render its offerings obsolete.
Revenue surge but volatile profitability
Fractal’s revenue increased by around 39% during FY23-25, from ₹1,985 crore in FY23 to ₹2,765 crore in FY25. The company reported ₹1,559 crore in revenue in H1FY26 and is expected to surpass FY25 revenue. Operating profitability, however, shows no clear trend. Ebitda declined from ₹437 crore in FY23 to ₹97 crore in FY24, and then recovered to ₹398 crore in FY25.
Why margins are under pressure
H1FY26 Ebitda rose to ₹186 crore from ₹131 crore in the same period last year. Margins follow a similar trajectory, declining from 22% in FY22 to 12% in H1FY26. This margin compression is primarily driven by the company's heavy reliance on human capital. Employee benefit expenses are Fractal’s largest expense, accounting for approximately 72% of total revenue.
To maintain its edge as a leading AI provider, the company maintains a highly selective workforce of 5,722 full-time employees, drawn from a pool with a 1.2% selection rate (964 hires from nearly 80,000 applications in H1FY26). However, intense competition for specialized AI talent and failure to retain it are direct threats to profitability.
Because Fractal’s top 10 client relationships average over eight years, the company may have limited pricing power to pass on these rising talent costs to its customers. Net profit is volatile, too. After reporting a profit of ₹194 crore in FY23, it posted a loss of ₹55 crore in FY24 and then reported a profit of ₹221 crore. In H1FY26, the company's net profit stands at ₹71 crore, down from ₹73 crore in H1FY25, and is likely to miss last year's number.
Next year, profitability could benefit as the company plans to utilize ₹265 crore of IPO proceeds to repay most of its outstanding debt of ₹275 crore. Given the interest expense of ₹23.3 crore in H1FY26, which annualizes to ₹46.6 crore, this prepayment should directly support the bottom line.
Net cash flow from operating activities is also volatile. It's down to 21.4 crore in H1FY26, from ₹40 crore in FY25, and ₹160 crore in FY24. At a market capitalization of ₹15,480 crore, the company is asking for a valuation of around 70 times price-to-earnings, which appears steep even after factoring in a scarcity premium.
A demanding price for scarcity
Although this valuation is 26% lower than the ₹20,978 crore valuation at which it last raised funds in July 2025, the multiples remain high, particularly given the inconsistent profitability and margin. Key-man risk remains material for Fractal as the business is deeply reliant on its two co-founders. The risk of losing either founder can materially threaten future operations and strategy.
For more such analysis, read Profit Pulse.
Madhvendra has over seven years of experience in equity markets and writes detailed research articles on listed Indian companies, sectoral trends, and macroeconomic developments.
The writer does not hold the stocks discussed in this article.
The purpose of this article is only to share interesting charts, data points, and thought-provoking opinions. It is NOT a recommendation. If you wish to consider an investment, you are strongly advised to consult your advisor. This article is strictly for educational purposes only.