Markets · IPO Analysis

The Company That Isn't Selling

OYO's zero-OFS IPO, and the case for a software business wearing a hotel uniform.

A reading of Oravel Stays' ₹6,650 crore fresh issue: what the absence of an Offer for Sale signals, how an asset-light, AI-run platform earns software-like margins inside hospitality, and where the real risks sit. Every figure is sourced to the company's Updated Draft Red Herring Prospectus-I (UDRHP-I, 29 June 2026) or to named public filings.

↳ Verify as you read: click any p.xxx citation to open the DRHP in a side drawer at that exact page, with the verbatim prospectus text shown in the drawer header. Hover a citation for the same text; press Esc or ✕ to close the drawer.

A large IPO where nobody sells

Oravel Stays Limited — the company most Indians still call OYO, now filing under the brand PRISM — is going public with a 100% Fresh Issue of ₹6,650 crore and an Offer for Sale of exactly nothing p.126. SoftBank, which holds 40.04% of the company pre-issue through SVF India Holdings (Cayman), is not selling a share. Founder Ritesh Agarwal, who controls 26.71% (6.59% directly and 20.12% through RA Hospitality Holdings, Cayman), is not selling a share. Airbnb, a 1.22% holder, is not selling either — confirmed by Bloomberg, The Week and Inc42.

That is rare, and the rarity is measurable. We built a dataset from the Securities and Exchange Board of India's own draft-offer archive — 4,672 draft red herring prospectuses filed between 2003 and 2026 — and classified each by whether it carried an Offer for Sale: a "cash-out" leg where existing shareholders sell into the IPO rather than the company raising fresh money. Of the filings we could confidently classify in 2021–2026, roughly 72% carried an Offer for Sale component; in 2021 the figure was 95%. For half a decade, the Indian IPO has functioned largely as an exit event, with the company itself often collecting only a slice of the headline number.

OYO has inverted that. Every rupee raised goes to the company, and the two shareholders best placed to sell — a founder and a global fund that together own two-thirds of the register — are choosing to stay whole. When the most-informed insiders keep all their capital in the business and route the proceeds into the balance sheet rather than their own pockets, the structure is telling you something about conviction. The rest of this analysis tests how much that conviction is worth.

What the money actually does

The proceeds are not a growth war-chest, and it matters to say so plainly. The Objects of the Issue p.293 allocate the net proceeds two ways: ₹4,987.5 crore (≈75% of gross proceeds) flows to the subsidiary Oravel Stays Singapore to repay or prepay borrowings — part of the Term Loan B (TLB), a five-year, ~US$830 million secured facility from Deutsche Bank's New York branch p.294 — and the balance, capped at 25%, funds general corporate purposes.

This is a deleveraging raise, and that is a feature. OYO borrowed roughly US$830 million in December 2024 to acquire G6 Hospitality — the franchisor of the Motel 6 and Studio 6 brands in the United States — for an all-cash ₹4,274 crore p.302. The IPO converts that acquisition debt into equity. In an Offer for Sale, cash leaves the ecosystem into a seller's account and the balance sheet is unchanged; here, ₹4,987.5 crore of debt comes off the consolidated books, and the benefit accrues to the continuing shareholder — the person buying in — not to an exiting one.

The repair is already visible in the numbers. Net leverage (net borrowings to EBITDA-excluding-exceptionals) has fallen from 12.28× in FY2023 to 3.08× (FY24), and stood at 2.60× for the nine months to December 2025 p.436. Finance costs still ran roughly ₹1,089 crore in nine months p.131, which is precisely why retiring the loan is the highest-return use of the money on offer. The honest summary is that all of the capital stays inside the company, and three-quarters of it repairs a balance sheet stretched by an acquisition already completed.

The business underneath the brandA software company wearing a hotel uniform

OYO does not own hotels. The prospectus is explicit that it operates "an asset-light business model and do[es] not own the storefronts listed on our platform" p.427. What it owns is a technology platform and a portfolio of brands, and it earns money four ways: revenue-sharing with Patrons (the hotel owners/operators), commission on home and listing bookings in Europe, royalty and franchise fees from G6 operators in the US, and subscription fees from European listings that pay to sit on the platform p.427.

Revenue-share, commission, royalty, subscription — that is the income statement of a platform, not a hotelier, and the mix is tilting further that way. Commission from bookings and royalty income reached ₹2,215 crore of the ₹6,941 crore of nine-month FY26 revenue, about 32%, up from roughly 22% in FY2023 p.436. The fee line is compounding faster than the room-nights line.

The technology is productised and named: PRISM OS for hotel operations, Grow for revenue management, Co-PRISM as the proprietary revenue stack, Atlas for property lifecycle management (it auto-generates digital contracts from KYC documents and handles e-signing), Bolt.ai as an in-house channel manager wired into more than 230 OTAs, metasearch platforms and global distribution systems, and PRISM Collect/Secure for payments and reconciliation p.441. The stack runs on cloud servers "enabling rapid iteration… with minimal marginal costs" p.438 — the defining economic property of software.

The margin proves the point. OYO reports Gross Profit of 19.19% of Gross Booking Value in FY2025 — roughly 1.5 to 5 times that of other leading Indian online travel agents, and the highest gross margin among India-listed hospitality-technology peers per its commissioned 1Lattice report p.431. A services company does not out-earn its peers' gross margin by multiples; a software layer sitting on a services industry does. The platform pays off for the other side too: hotels on it see a 1.16× to 2.27× uplift in net realised value after joining p.428, and over 65% of Patrons who joined between April 2023 and December 2025 had no bookable online presence with any leading OTA beforehand p.439. OYO is not dividing an existing distribution market; it is manufacturing distribution where none existed — the market-expanding move of a software business.

Even the cost structure reads like software: 76.12% of employees sit in India p.427, the engineering base for a company whose revenue is now majority-international. Build the software cheaply in India; deploy it at software margins across the world.

The surface is hospitality — rooms, check-ins, brands. The economics are software: recurring fees, multiples-of-peers gross margin, minimal marginal cost, an India engineering base serving global revenue.

The Sequoia lens

In March 2026, Sequoia Capital's Julien Bek published "Services: The New Software," arguing that "the next $1T company will be a software company masquerading as a services firm." The reasoning turns on an asymmetry — for every dollar the world spends on software, it spends roughly six on services — and on a distinction between two kinds of AI product. A copilot sells a tool to a professional who still does the work; an autopilot does the work and bills for the outcome. The autopilots win, Bek argues, because selling a tool puts you in a race against the model, while selling the work means every model improvement makes your service cheaper and harder to displace.

OYO sits on the autopilot side of that line more than the copilot side. Its pricing engine does not hand hoteliers a dashboard and wish them luck; it sets the prices itself, allowing owners only "a limited degree of flexibility… within a specified range (rather than allowing Patrons to set their own prices manually)" p.440, runs the revenue-management, distribution and operations functions through PRISM OS, and keeps a share of the outcome. It sells the work of running a hotel's commercial engine, not merely the tool.

Here is where I will put a view rather than a fact. OYO's model is not finished — it is early, and that is the opportunity. The company has spent the last three years doing the hard, unglamorous part: rebuilding the platform, culling unprofitable supply, and buying a mature US fee-stream. What it has not yet done is run that machine at full optimisation with AI as the primary growth input. It does not need a fresh mountain of capital to do so. Once ₹4,987.5 crore of debt is retired, the roughly 25% of proceeds reserved for general corporate purposes — on the order of ₹1,600 crore — sits alongside operating cash flow that already reached ₹1,594 crore in nine months p.133. On an asset-light platform, growth is not paid for in property; it is paid for in software and talent — both of which OYO produces at an India cost base. That is why the deleveraging is not a defensive crouch: it clears the balance sheet so that internally generated cash and a modest reserve can compound through AI-driven margin expansion, rather than being consumed by interest. In my reading, the next leg of this company's growth is a margin story more than a capital story, and it is precisely the kind of shift Bek describes — a services business quietly re-rating toward software economics.

The AI wedge

The strongest part of the forward case is that OYO owns the two inputs applied AI is starving for — proprietary operating data and direct distribution — and owns them in a form its competitors do not.

The prospectus already describes an AI layer doing real work: models that process platform interactions to power "personalized recommendations, real-time dynamic inventory pricing, automated cataloguing and content enhancement, fraud and risk controls and intelligent service" p.438. The dynamic-pricing engine recalibrates rates hourly across demand, seasonality, competitor pricing, weather, local events and cancellation signals p.440. There is an AI voice-and-chat bot for guests and partners, an automated refund system, and automated contracting through Atlas.

Why this compounds specifically for OYO, and not for a pure listings marketplace: a conventional OTA sees only what it lists — search, click, book. OYO operates the storefront, so it captures the far richer downstream data of running the asset — realised occupancy, price-response curves, cancellations, guest behaviour, on-property outcomes — across 293,554 storefronts and 119.36 million customers since 2012, with 67.57% of demand arriving directly through its own channels rather than third-party OTAs p.430. That is a proprietary training set an inventory aggregator cannot assemble, because the aggregator never runs the hotel. Feed it into an autopilot that already sets prices and manages operations, and each model improvement lifts the margin on the entire network at once — with the marginal cost of applying it to one more hotel close to zero p.438.

The mechanics of AI-led margin expansion are concrete rather than speculative. On the demand side, better ranking, imagery and personalisation raise conversion and pull more traffic through owned channels, which lowers customer-acquisition cost and lifts the 67.57% direct-demand share further. On the supply side, automated cataloguing, AI resolution of refunds and queries, and algorithmic operations strip cost out of functions that a traditional hotel company staffs with people. On price, an engine recalibrating hourly captures yield that manual pricing leaves on the table. None of these require OYO to own a single additional room. On an asset-light base, AI is not one growth lever among many — it is the principal one, because the alternative levers (building or buying property) are exactly what this model was designed to avoid. That is the sense in which OYO is unusually well positioned to let AI, not capital, drive its next phase.

One model, many countries

The international footprint is the proof that the software travels. OYO calls the approach "India for the World" p.427: build the technology, brand and operating playbook in India, then replicate it across markets with little local reinvention. The business runs across three verticals — Hotels (Sunday, Townhouse, Palette, OYO, and now Motel 6/Studio 6), Homes (Belvilla, DanCenter, Checkmyguest in France, MadeComfy in Australia/NZ), and Listings (European storefronts on a fixed subscription) — spanning more than 35 countries p.427.

The G6 acquisition has redrawn the map. For the nine months to December 2025, Gross Booking Value by geography ran United States ₹12,023 crore, Southeast Asia/Middle East & International ₹4,153 crore, Europe ₹4,039 crore, and India ₹2,732 crore p.435 — meaning the United States is now OYO's single largest booking geography, a striking outcome for a company founded in Gurugram. Crucially, G6 is a franchising and fee business, not real estate: OYO bought the franchisor that collects royalty and marketing fees from independently owned Motel 6 and Studio 6 properties, a stream that generated ₹4,642 crore of revenue in the April–December 2025 period p.302. Layered onto an India-built platform and India cost base, that is exactly the kind of asset the "India for the World" model is designed to monetise. The one uneven edge is distribution maturity: in Europe, 61.78% of used room nights still arrive via third-party OTAs versus 27.54% in the US p.431 — the direct-demand flywheel that works so well in India is not yet uniform across the map, which is both a risk and the headroom.

The turnaround, read honestly

Revenue is re-accelerating — up 16% in FY25, with nine-month FY26 already exceeding the full prior year — and the clean operating metric, EBITDA excluding exceptionals, share-based pay and other income, has climbed from ₹274 crore (FY23) to ₹898 crore, ₹1,095 crore, and ₹1,968 crore in nine months p.436. Operating cash generation of ₹1,594 crore in nine months is the most reassuring line in the accounts.

Two caveats belong next to that, and leaving them out would weaken the piece, not strengthen it. First, FY2025's reported net profit of ₹245 crore was not an operating profit: profit before tax that year was negative ₹489 crore, and the positive bottom line came from a ₹768 crore deferred-tax credit p.131. The nine-month FY26 figures are cleaner, with profit before tax genuinely positive. Second, the take rate is compressing — revenue as a share of GBV has fallen from 53.7% (FY23) to 30.25% (9M-FY26) p.436 — as the mix shifts to lower-take, higher-margin franchise and US business. Gross-profit margin on GBV is broadly holding, so this is mix, not decay; but it means headline GBV growth flatters revenue growth.

The evidentiary core: how OYO's structure compares

Even trending down — from 95% in 2021 to 59% in 2025 — the pattern is unambiguous: for most of the last five years, the majority of Indian IPO filings carried a cash-out leg. Against every one of those baselines, OYO's zero is an outlier. The marquee comparison, drawn from public filings, shows the full spread:

The spread is the point. Not every new-economy IPO was a cash-out — Zomato was 96% fresh — but the bulk of the cohort (Paytm, Swiggy, Nykaa, LIC) raised most of their money for selling shareholders. OYO stands at the far end: not merely fresh-heavy but zero OFS, at ₹6,650 crore of scale, with its two largest owners contractually staying in. Among large Indian new-age listings, that combination is close to unique.

Zero-OFS does not prove the shares are cheap. Valuation will be set by the price band, which the prospectus leaves blank p.120, and a strong structure at the wrong price is still a poor entry. What it does provide is alignment: the best-informed holders are keeping their capital in the business, and the proceeds strengthen the company rather than enriching an exit. In a market where the reverse is the default, that is information worth paying attention to.

Why the risks are the price of the position

The bear case is real and belongs on the table, drawn from the prospectus's own Risk Factors p.31 and financials:

• Leverage still bites. Even after the planned repayment, OYO carried roughly ₹8,328 crore of fund-based borrowings as of 31 May 2026 p.299, and finance costs consumed ~₹1,089 crore in nine months. The equity case depends on the deleveraging actually completing.
• Goodwill and intangibles are heavy. The balance sheet carries ₹6,386 crore of goodwill and ₹5,342 crore of other intangibles p.129, mostly from acquisitions; if G6 underperforms, impairment can swing reported earnings.
• Profitability is young and partly tax-driven, as the FY25 deferred-tax credit shows.
• Execution risk on a US-centric pivot — the US became the largest GBV geography barely a year after the acquisition.
• Structural complexity — 114 subsidiaries p.121, possible US "passive foreign investment company" status for US holders p.122, and multi-jurisdiction regulation.
• Market-structure risks — no trading history, an undetermined price band, potential post-listing surveillance measures, and no dividend record p.120.
• History — OYO's private valuation was marked down sharply in prior years; the turnaround, while credible, is recent.

These are not reasons to look away — they are the shape of a company still building a category rather than harvesting one. Read them through Bek's lens and the calculus changes. The prize for "selling the work" is largest exactly where the underlying market is enormous, fragmented and unorganised — and hospitality's budget and economy segments are among the most unorganised markets on earth, with 60% to over 90% of supply outside any branded, technology-enabled system and OYO holding barely 1–2% of it today p.429. In a market that shape, the company that builds the operating system for budget supply and compounds it with AI does not win a slice — it becomes the default infrastructure, with the winner-take-most economics that follow scale, data and switching costs. The leverage, the goodwill, the youth of the profit line are the cost of assembling that position now, while the field is still open, rather than renting it later from whoever gets there first. The bet is not that OYO is cheap next quarter; it is that OYO is playing a decade-long game in which AI widens the moat every quarter, and that the entry point is a company raising money to build, not to sell.

The bottom line

What OYO is bringing to market is a structure the Indian IPO rarely offers: a large fresh issue with zero cash-out, owners who are staying whole, proceeds that repair rather than reward, and — underneath the hotel branding — the economics, cost base and data assets of a software company. That combination is unusual on its own. What makes it interesting is the direction of travel.

The opportunity in front of this company is not incremental. It is one of the largest unorganised services markets in the world, barely penetrated, being pulled online for the first time — and OYO is the rare operator that both runs the supply and owns the demand, which is the exact position from which AI produces compounding margin rather than cosmetic features. If the next few years play out on the trajectory the numbers already trace — revenue re-accelerating, leverage falling, the fee mix rising, EBITDA-excluding compounding, and AI turning proprietary operating data into yield and cost-out that no listings marketplace can copy — then OYO does not end up as a better hotel aggregator. It ends up as the operating system for the world's budget and economy hospitality, earning software margins on a services-scale market, with the United States already its largest geography and the India cost base funding the whole thing. That is a genuine shot at category-defining, winner-take-most infrastructure, and it is being underwritten by insiders who are buying the future with their own capital, not selling it.

None of that is a promise, and one discipline stays in force until the RHP fills in the blanks: the price band will decide how much of this future the market is asking you to pay for up front, and the durability of the clean operating profit is the number to keep watching. But on the evidence in its own prospectus, this is not the ordinary Indian listing — a partial exit dressed as a fundraising. It is the rarer thing: a company raising money to build and repair itself, run by people staying in their seats, at the front edge of a market large enough, and unorganised enough, to be worth building an operating system for.

Disclaimer: This article is journalism and analysis, not investment advice, and its author is not a registered investment adviser. It concerns a proposed public offering whose price band, final structure and regulatory clearance were not settled at the time of writing; a draft red herring prospectus is not an invitation to invest, and the equity shares referred to have not been recommended or approved by SEBI. Nothing here is a recommendation to buy, sell or subscribe to any security. Figures are drawn from the company's draft prospectus and named public sources and may be revised in the final prospectus. Readers must read the Red Herring Prospectus in full — including its Risk Factors — and take independent professional advice before making any investment decision.