CleanMax: India's Bet on Corporate Clean Power
I. Introduction & Episode Roadmap
On the morning of March 2, 2026, a group of bankers gathered in a conference room in Mumbai's Bandra Kurla Complex to watch a listing they had spent eighteen months building toward. The stock opened at ₹960 against an issue price of ₹1,053 — a 9% discount, ugly but survivable.1 Then it kept falling. By mid-morning the screens showed a drawdown of 28%. When the closing bell rang, Clean Max Enviro Energy Solutions had settled at ₹867.50, down 17.6% on debut.2 For an offering that raised $341 million, it was the worst first-day performance by any Indian IPO of that size since Paytm's infamous 2021 flop — a comparison that no management team wants attached to its listing day.2
Four months later, the same stock traded at ₹1,400.3 It had more than doubled from its March low of ₹727, touched a high above ₹1,530 in late June, and by August 12, 2026 sat at ₹1,266 — comfortably above its issue price, with a market capitalisation of roughly ₹14,800 crore (about $1.7 billion).4 Nothing about the underlying business had changed direction in those four months. The plants under construction in March were the same plants commissioned in June. The contracts signed in 2024 and 2025 were the same contracts generating revenue in 2026. What changed was the market's willingness to pay for a story it had, briefly, refused to buy.
That is the puzzle worth sitting with. Because CleanMax is not a consumer brand or a software platform whose value is genuinely hard to pin down. It is an infrastructure company. It builds solar farms and wind turbines, signs fifteen-year contracts with factories and office parks and — increasingly — with the data centres running India's artificial intelligence build-out, and it sells them electricity at a discount to what the state utility would charge. The cash flows are contracted. The assets are physical. And yet the market managed to price this business at ₹727 and ₹1,533 within a single quarter.
This is the story of how a McKinsey partner's failed gas-power venture became India's largest supplier of clean electricity to corporate customers; how a Canadian infrastructure giant wrote a cheque that changed the company's ceiling; how demand from Google, Meta, Amazon, Apple and a dozen data-centre operators went from a rounding error to 42% of the contracted book in about two years; and why the same asset-heavy, debt-funded model that produced 59% operating margins also produced a balance sheet that grew borrowings nearly tenfold in three years.
It is also, unavoidably, a story about what the evidence does and does not support. CleanMax's management makes a specific claim: that fifteen years of structuring expertise in one of the world's most fragmented and regulated power markets constitutes a durable advantage that better-capitalised rivals cannot easily replicate. That claim deserves testing rather than repeating — and by the end of this piece, the reader should have a clear view of what would confirm it and what would falsify it.
Start where the company started: with a business plan that died within a year.
Subscribe on LinkedInII. Origins: Why a McKinsey Partner Went Into Solar (2010–2015)
In 2011, Kuldeep Jain was thirty-six years old and had just walked away from roughly twelve years at McKinsey & Company, where he had risen to global partner leading the firm's energy and corporate finance practices.5 He had spent more than a decade advising other people on how to allocate capital in the power sector. He decided he wanted to do it himself.
His first business plan was not solar. It was gas.
The logic was impeccable on paper. India in 2010 had a chronic electricity problem: industrial and commercial users paid some of the highest power tariffs in the world, cross-subsidising agricultural and residential consumers through a tariff structure that state distribution companies — the discoms — had used for decades. Those same industrial users endured load-shedding, voltage fluctuation and unpredictable supply, and many ran diesel generators as backup at costs that made grid power look cheap by comparison. Gas-fired generation, clean-burning and quick to build, looked like the obvious arbitrage.
Then, in March 2011, an earthquake and tsunami hit the Fukushima Daiichi nuclear plant in Japan. Japan idled its nuclear fleet and went shopping for liquefied natural gas. Global LNG prices roughly doubled. India's gas-fired generation economics collapsed, and a fleet of gas plants across the country stranded. CleanMax's business plan died with them, inside its first year.
Jain has described that moment with unusual bluntness: "We came to a point where we suddenly had no business plan. I had put in a lot of money and it was also running dry."5
What happened next is the actual founding story. In 2012, Jain pivoted the company — then called CleanMax Solar — into rooftop solar.5 Solar module prices were entering the collapse that would define the decade, Chinese manufacturing capacity was flooding the market, and the levelised cost of a rooftop solar installation in a sunny Indian state was about to cross below what a factory paid its discom.
But the crossover alone was not a business. It was a spreadsheet observation available to anyone. The barrier was that the customer — a mid-size auto components maker in Pune, a pharmaceutical plant outside Ahmedabad — did not want to become a power producer. Solar was a capital expenditure they did not understand, with a technology risk they could not evaluate, an operations burden they did not want, and a payback period their CFO would not approve when the same rupee could buy an extra production line.
Jain's answer was a financing structure, not a technology. CleanMax would build the plant on the customer's roof, own it, finance it, insure it, maintain it and operate it. The customer would sign a long-term power purchase agreement — a PPA — and pay only for the electricity actually generated, typically 30–40% below the prevailing grid tariff.5 Zero upfront capital. No technology risk. No operating headache. If the panels underperformed, CleanMax ate the loss.
In India this became known as the opex model, as distinct from the capex model where the customer buys the plant outright. CleanMax did not invent the concept globally — American residential solar had been doing third-party ownership for years — but it was among the first to industrialise it for Indian corporate customers, and it built the underwriting muscle to do it at scale: credit-assessing hundreds of counterparties, navigating state-by-state approvals, and structuring around the fact that Indian electricity regulation is not one market but roughly thirty.
The strategic elegance is worth stating plainly, because it explains everything that follows. By taking the balance sheet burden onto itself, CleanMax converted a capital-allocation decision — which every corporate CFO would defer — into a procurement decision, which a plant manager could sign. That is a distribution unlock, and distribution unlocks tend to compound faster than technology ones.
The traction followed the structure. By the early 2020s the company counted more than 200 corporate customers, including Facebook, Tata Motors, Grasim and Cipla, with over 550 rooftop installations and more than 250 MW of capacity, and revenue that grew from ₹319.8 crore in FY20 to ₹637.2 crore in FY21.5 Bridge to India's rankings placed CleanMax at the top of the country's rooftop solar developer league tables for several consecutive years. In 2019, the company extended the same opex logic beyond rooftops into wind and wind-solar hybrid projects — a move that mattered more than it appeared, because rooftops are physically capped by available roof area, and a factory that wants 100% clean power needs far more electricity than its own roof can produce.5
That constraint — roofs are small, corporate ambition is not — is what pushed CleanMax off the roof and into the far more capital-hungry business it operates today. As Jain put it, corporates "don't talk 20-30 percent green, they want 100 percent green."5 Meeting that demand required something the founder could not supply on his own: several billion dollars of patient capital.
III. The Business Model: Two Segments, One Flywheel
Picture a pharmaceutical plant on the outskirts of Hyderabad. It draws 12 MW around the clock. Its rooftop, even fully covered in panels, might yield 3 MW at peak sun and nothing after dark. To get to meaningful renewable coverage, the electrons have to come from somewhere else — a solar farm in Rajasthan, a wind site in Karnataka — and travel to the plant across transmission lines the company does not own, under rules written by a state regulator that changes them periodically.
Solving that problem, repeatedly, across hundreds of customers and roughly a dozen state regulatory regimes, is what CleanMax actually sells. The electron is a commodity. The structuring is not.
The company reports in two segments. The larger one, Renewable Energy Power Sales, accounted for roughly three-quarters of FY25 revenue and is the engine of the business.6 Here CleanMax develops, owns and operates generation assets — rooftop arrays, ground-mounted open-access solar farms, wind farms and hybrids — and sells the output to corporate customers under long-term PPAs. Some of these run through "open access," where a generator wheels power across the grid to a specific consumer and pays transmission and surcharge costs. Others use the "group captive" structure, where the customer takes at least a 26% equity stake in the generating entity and consumes at least 51% of its output, thereby qualifying as a captive user and becoming exempt from the cross-subsidy surcharge and additional surcharge that discoms levy on open-access consumers.7 That surcharge exemption is often the difference between a PPA that clears a customer's hurdle rate and one that does not — which means the structure, not the sunshine, frequently determines whether the deal happens.
It is worth pausing on the plumbing, because "open access" is the term that does most of the work in Indian C&I renewables and is rarely explained. A solar farm in Rajasthan cannot run a private wire to a factory in Tamil Nadu. Instead it injects power into the grid and the factory withdraws an equivalent amount, with the transmission network acting as a shared pipe. The generator pays wheeling charges for using the pipe, transmission charges for the long-haul portion, and — unless exempt — a cross-subsidy surcharge compensating the discom for the profitable industrial customer it just lost. A related mechanism called banking lets a generator inject surplus during sunny hours and withdraw the credit later, effectively using the grid as a battery; states set the rules on how much can be banked, for how long, and at what charge. Every one of those line items is set by a state regulator, varies across the country, and gets revised. The delivered cost of a clean electron in India is therefore mostly a regulatory outcome, not an engineering one — which is exactly why a company whose competitive claim rests on structuring expertise can exist at all.
The smaller segment, Renewable Energy Services, contributed roughly the remaining quarter of FY25 revenue.6 This is turnkey engineering, procurement and construction for customers who prefer to own their assets outright, plus operations and maintenance, and an assortment of ancillary offerings: renewable energy certificates, carbon credits, single-axis trackers, robotic panel cleaning, remote monitoring. Services is a useful complement — it keeps CleanMax in the room with customers who are not ready for a PPA, and it generates cash without consuming much balance sheet — but it is not where the value sits, and this article will size it accordingly.
The economic distinction between the two segments explains why CleanMax looks so unusual on a margin line. An EPC contractor books the full project cost as revenue and earns a single-digit or low-teens margin on it. An asset owner books only the electricity sold, at a much smaller revenue number, but keeps most of it as gross profit because sunlight and wind have no fuel cost. In FY26, CleanMax reported operating profit of ₹1,132 crore on revenue of ₹1,913 crore — an operating margin of roughly 59%.8 On the company's own adjusted EBITDA definition, the FY25 margin was closer to 68%, against a reported operating margin of 60% on the same year's revenue.98 The gap between those two numbers is not fraud; it is the ordinary result of different treatment of other income and one-off items. But it is a reminder to any investor reading company presentations that "adjusted EBITDA" in an infrastructure business is a management-defined metric, and the reported operating line is the one to anchor on.
The more important point about those margins is what they are not. A 59% operating margin in a manufacturing business would signal enormous pricing power. Here it signals almost the opposite: it is an artefact of an asset-heavy structure where the cost is capital, not operations. Below the operating line, FY26 interest expense was ₹786 crore and depreciation ₹380 crore.8 Together those two lines consumed the overwhelming majority of the operating profit, leaving ₹86 crore of net profit. This is the defining feature of the business model, and it deserves to be stated without euphemism: CleanMax is a leveraged infrastructure owner. Its margins are high because its capital intensity is high. The two are the same fact viewed from opposite ends.
What makes the C&I niche worth occupying is the price realised per unit. Antique Stock Broking, initiating coverage in July 2026, estimated CleanMax's EBITDA realisation at roughly ₹3.85 per kWh, compared with about ₹2.50 per kWh for utility-scale capacity won in competitive central auctions.10 That spread is the entire commercial case for specialising in corporates. In a reverse auction against NTPC or SECI, a developer bids against a dozen rivals for an anonymous offtake contract, and the lowest bid wins on cost of capital alone. Selling to a factory is a bilateral negotiation, priced not against another developer's bid but against the customer's alternative — the discom tariff, which is high precisely because of the cross-subsidy structure that makes Indian industrial power expensive.
The flywheel, to the extent one exists, works like this: a customer starts with a rooftop installation, discovers the savings are real and the operator competent, and comes back for open-access capacity when it wants deeper decarbonisation. Antique counted 588 corporate customers across 1,280 individual PPAs, with roughly 74% repeat business and more than 95% of the contracted book rated A- or better on domestic credit scales.10 In the June 2026 quarter, the company said 79% of new capacity additions came from existing customers.11
That last statistic is the one worth interrogating. A high repeat rate can mean two very different things: that customers are locked in and delighted, or that CleanMax's sales motion is efficient at farming a base it already has while struggling to break into new accounts. The credit quality of the book and the absence — per Antique's diligence — of any contract renegotiation against the company since inception argue for the benign reading.10 But a PPA is a fifteen-to-twenty-five-year contract with a defined end, and the switching cost at renewal is genuinely low: the plant is depreciated, the customer knows the market, and a rival can bid for the renewal with a fresh cost base. Nobody has yet observed a large CleanMax renewal cohort come up for repricing at scale. That is an open question, not a settled advantage.
For investors, the model resolves to a simple frame: this is a spread business. CleanMax borrows at one rate, builds assets at one cost per megawatt, and contracts them at one tariff. Everything that follows in this story — the Brookfield deal, the data-centre boom, the margin compression, the leverage debate — is a question about whether that spread holds as the business scales.
IV. The Brookfield Deal: The Inflection Point (2023)
By early 2023, CleanMax had a good problem. It operated more than 1.6 GW of wind and solar assets and had a pipeline of corporate demand it could see clearly.12 What it did not have was the capital to serve that demand. Every incremental megawatt required roughly seventy to eighty percent debt and twenty to thirty percent equity, and the equity had to come from somewhere. Founder-controlled companies backed by growth-stage private equity — CleanMax had raised roughly $175 million cumulatively by 2022 — do not build gigawatt-scale infrastructure platforms.5
On June 2, 2023, Brookfield Renewable announced it had acquired a controlling stake in CleanMax through the Brookfield Global Transition Fund, in a $360 million equity investment combining primary and secondary capital.12 Jain's framing at the time was pragmatic to the point of being unromantic: "This will see us have adequate growth capital for at least the next 3-4 years," he said, adding that with Brookfield's capital pool the company expected to add roughly 800–1,000 MW annually to its C&I portfolio and become a platform of more than 5 GW within three to four years.12
Read that target again with the benefit of three years' hindsight, because it is the single most useful test of management credibility available in this story. In June 2023, the company said it would reach 5 GW-plus in three to four years and add 800–1,000 MW a year. By March 2026 the contracted portfolio stood at roughly 5.7 GW, with about 1.4 GW commissioned during FY26 alone.13 By June 30, 2026, contracted capacity across both segments reached 6.8 GW.11 Management set a public, specific, multi-year target under a new controlling shareholder and hit it early. That is a real data point, and one that most Indian infrastructure companies cannot produce.
What Brookfield brought was not simply money. It brought a specific kind of money. Brookfield is one of the world's largest infrastructure asset managers, with a cost of capital and a duration tolerance that venture and growth-equity investors do not possess. An infrastructure fund underwrites twenty-year contracted cash flows at leverage levels a generalist would find alarming, because that is what the asset class supports. Attaching that balance sheet to a company that had proven it could originate corporate PPAs was the actual transaction: origination capability married to permanent capital.
It also changed who CleanMax was. It became a co-promoted company — Brookfield alongside Jain — a structure that is common in Indian infrastructure but still unusual for a founder-led business going public. Around this period the shareholder register accumulated an international cast: Augment Infrastructure, an anchor private equity backer since the 2018–19 period, and Danish development-finance capital through the IFU-linked vehicle. Then, on February 6, 2026, barely two weeks before the IPO opened, the company raised ₹1,500 crore in a pre-IPO round from Temasek Holdings, Bain Capital, 360 One, the Steinberg India Emerging Opportunities fund, Steadview Capital, and Indian family offices including the Dalmia group and the Jaisinghani and Taparia families.9
Pre-IPO rounds that large, that close to a listing, cut both ways analytically. The bullish reading is that sophisticated long-duration investors — Temasek in particular, which then also anchored the IPO itself — did the diligence and committed. The skeptical reading is that a company confident of strong public demand does not need to place ₹1,500 crore privately seventeen days before its book opens. Given what subsequently happened to the order book, the skeptical reading has some support.
The "co-promoter" designation is not cosmetic. Under Indian securities regulation, promoter status carries specific obligations — lock-in requirements on shares at IPO, disclosure duties, abstention from voting on related party transactions, and a degree of accountability for the company's conduct that an ordinary large shareholder does not bear. When a global asset manager accepts that designation rather than structuring itself as a financial investor, it is accepting regulatory responsibility for the business, which is a stronger commitment than a board seat. It also means the founder no longer controls the company in the way Indian markets typically assume when they see a founder-led business.
Post-IPO, the promoter group holds roughly 49.4%, with foreign institutional investors at 11.2%, domestic institutions at 14.7%, and the public at 24.7% as of June 2026.8 Promoter holding fell from 74.89% pre-issue to 49.08% at listing.1 Brookfield remains the dominant voice in that promoter block.
The governance implications are worth naming rather than glossing. Minority shareholders in CleanMax are, in practice, riding alongside a private-market infrastructure investor whose fund has a defined life, a target return, and an eventual exit obligation to its own limited partners. That alignment is excellent while the strategy is "build aggressively at attractive returns" — Brookfield wants exactly what a growth-oriented public shareholder wants. It becomes more complicated later, when the fund's clock starts running and a large block has to find a home. This is not a criticism; it is a structural fact that determines whose timetable sets the agenda. Public investors in Brookfield-controlled listed vehicles have historically done fine, but they have rarely been the ones deciding when the story ends.
With permanent capital secured and a controlling shareholder that measures success in gigawatts, CleanMax spent the next three years doing the only thing that structure permits: building, at a pace that would have been unimaginable in 2022 — and running straight into a demand shock nobody had modelled.
V. The Scaling Era and the AI Power Story (2023–2026)
In March 2024, CleanMax's contracted book included 240 MW of capacity earmarked for data centres. It was a line item — real, growing, unremarkable. Twenty-seven months later, in June 2026, that figure had grown to more than 2.5 GW, representing 42% of the entire contracted RE Power Sales portfolio.11 Roughly a tenfold increase, in a segment that had barely registered when the Brookfield deal closed.
Understanding why requires understanding what changed in the physical world. A traditional enterprise data centre is a modest electricity consumer. An AI training facility is not. Racks that once drew 5 to 10 kilowatts now draw 40 to 100 or more when packed with GPU clusters, and unlike a factory that runs two shifts, these facilities draw close to their peak load twenty-four hours a day, every day. The useful analogy: a conventional data centre is a large office building; an AI data centre is an aluminium smelter that happens to run software.
India became a destination for that build-out for reasons that have little to do with clean energy — data localisation rules, a vast domestic user base, engineering talent, and land and construction costs far below Northern Virginia or Dublin. India added 258 MW of new data centre capacity in the first half of 2026 alone, up 59% from 162 MW in the same period of 2025.14 The hyperscalers building that capacity — Microsoft, Google, Amazon, Meta, Apple, and the colocation operators serving them — arrived carrying corporate net-zero commitments made in Redmond and Cupertino, commitments that apply to every megawatt-hour consumed anywhere on earth.
This produced a specific, acute scarcity: bankable, contractable clean power in India, at scale, delivered to a specific site, under a contract structure that a Fortune 100 procurement team and its auditors would accept. Indian discoms could not supply it. Utility-scale developers selling into central auctions were not structured for it. CleanMax had spent fifteen years building exactly that capability for exactly that kind of customer, just at a smaller unit size.
The customer roster reflects the shift. CleanMax's data-centre and technology clients include Equinix, STT Global Data Centres, NTT, Iron Mountain, Princeton Digital Group, L&T Data Centre and Cisco.11 Equinix signed its first India PPA with CleanMax in November 2024, a 33 MW captive solar-wind project in Maharashtra intended to take its Mumbai facilities to full renewable coverage. And in May 2026, Apple announced a co-investment partnership with CleanMax, committing an initial ₹1 billion (roughly $10.6 million) toward developing more than 150 MW of new renewable capacity in India — an extension of a 2024 arrangement under which CleanMax already powered Apple's Indian offices and retail stores.1516
The Apple structure deserves a moment. Apple is not merely signing a PPA; it is co-investing equity in the projects. That is a meaningfully different relationship — it lengthens the tie, aligns the counterparty to project success rather than just tariff, and signals a level of diligence on CleanMax's operating capability that a procurement contract alone would not. Antique's estimate is that CleanMax has captured something like 35–38% of India's hyperscaler clean energy deals.10
There is a technical wrinkle here that shapes the economics and is often glossed over in the AI-power narrative. A data centre needs power every hour of every day. Solar produces for roughly a third of the day; wind produces intermittently and seasonally. Neither, alone, matches a data centre's load curve. The practical solution CleanMax and its peers use is a hybrid portfolio — solar and wind sited in complementary resource zones, sized deliberately larger than the customer's requirement, combined with grid banking and, increasingly, batteries. The result is typically 50–70% renewable coverage on an annualised basis rather than genuine hour-by-hour matching, with the grid supplying the balance.
That matters in two directions. It caps how much of a hyperscaler's load any developer can claim today, which is a ceiling on near-term revenue per customer. It also creates the next round of demand: as customers push from annualised matching toward round-the-clock clean supply, the megawatt-hours required per customer rise substantially and storage becomes part of the contract. Whether CleanMax can profitably deliver firm, storage-backed power is an open question — batteries add capital cost with no fuel savings to offset it — but it is the direction the market is moving, and it is where the next competitive test will be fought.
The operational response has been the fastest build-out in the company's history. FY26 saw roughly 1.4 GW commissioned, taking operational capacity to about 3.1 GW — an 80% year-on-year increase.13 The June 2026 quarter delivered a record 0.5 GW of commissioning in three months, taking operational capacity to 3.5 GW with a further 2.5 GW under execution.11 Along the way the company commissioned its first 525 MWp project connected to the central transmission network, in Bikaner, Rajasthan — a technical step-change from rooftop arrays measured in kilowatts to grid-scale assets feeding the interstate network.13 Management has guided to a minimum of 1.5 GW of additions in FY27.11
Underneath the headline numbers, the customer base remains genuinely broad: 593 commercial and industrial clients as of June 2026, spanning automotive, pharmaceuticals, FMCG, IT services, textiles, chemicals and digital infrastructure.11 No single sector other than data centres dominates, and the long tail of mid-size Indian manufacturers is what generated the original relationships.
But the concentration question is real and should not be waved away by pointing at 593 logos. Customer count measures relationships; megawatts measure economics. If 42% of contracted capacity sits with data-centre operators, and a large share of that sits with a handful of global hyperscalers and their colocation partners, then the revenue concentration is materially higher than the customer count implies. The bull case treats this as a moat: hyperscalers are the most creditworthy counterparties on earth, they sign the longest contracts, and once a developer is on an approved vendor list the relationship compounds. The bear case observes that the same handful of companies set global AI capex, that AI infrastructure spending has been revised in both directions before, and that a contracted-but-not-yet-built megawatt is not the same thing as a delivered one.
There is a third possibility worth holding, which is neither bullish nor bearish but structural: hyperscalers are sophisticated, repeat, high-volume buyers with in-house energy teams. Over time, sophisticated repeat buyers tend to compress their suppliers' margins. The pricing CleanMax achieves today reflects genuine scarcity of bankable clean power in India. If that scarcity eases — because Adani, ReNew, Tata Power and the rest all build for the same customers — the negotiating dynamic shifts. Concentration risk in this business may show up first as price, not as loss of contract.
Which raises the obvious question: what exactly stops the better-capitalised players from taking this business?
VI. Industry Structure & Competition: Why CleanMax Wins (or Doesn't)
To understand where CleanMax sits, picture the Indian renewables landscape as two adjacent businesses that look identical from outside and behave nothing alike.
The first is utility-scale. Its emblem is Khavda, a 538 square kilometre expanse of barren salt flat in Kutch, Gujarat — a footprint five times the size of Paris — where Adani Green Energy is building what it describes as the world's largest renewable energy plant, with 30 GW planned and 9.5 GW of solar commissioned so far.17 On July 1, 2026, Adani Green became the first Indian renewables company to cross 20 GW of operational capacity, comprising 14.2 GW of solar, 2.7 GW of wind and 3.3 GW of hybrid, generating roughly 3% of India's total electricity consumption.17 ReNew, listed on Nasdaq, commissioned 2.4 GW in FY26 to reach roughly 12.6 GW operational.18 Tata Power Renewable Energy held about 6.1 GW of utility-scale renewables as of the December 2025 quarter.19
This is a game of scale, land aggregation, transmission connectivity and — above all — cost of capital. Developers bid into reverse auctions run by central and state agencies. Contracts are twenty-five years and effectively risk-free on the offtake side. Margins are thin and the winner is whoever can fund cheapest.
The second business is commercial and industrial. Its emblem is not a salt flat but a spreadsheet: a bilateral negotiation with a corporate CFO over a tariff, a term, an escalation clause and a group captive structure that must satisfy a state regulator. Here the competitive set is different — CleanMax, Fourth Partner Energy, Amplus (now under Petronas's Gentari), Avaada Energy, AmpIn Energy Transition, O2 Power and Serentica. This is a game of origination, credit underwriting, regulatory structuring and servicing hundreds of small counterparties rather than a few large ones.
CleanMax's claimed edge lives entirely in the second game. Antique put its national C&I share at roughly 12%.10 Bridge to India's rankings have placed it first among rooftop solar developers for several years running. Those are real achievements, and they should be read for what they are: leadership of a fragmented market, not dominance of a concentrated one. A 12% share means 88% of the market belongs to someone else. It also means that if a customer decides CleanMax is too expensive, there are at least half a dozen credible alternatives with a sales team already calling.
Run Porter's five forces across this and the picture is honest rather than flattering. Buyer power is meaningful and rising: corporate customers negotiate bilaterally, hold competitive tenders, and — in the case of hyperscalers — employ energy procurement professionals who know exactly what a megawatt should cost. Supplier power is moderate: solar modules are a commoditised, oversupplied global market, which helps, but turbines are more concentrated and the real supplier here is the capital market, whose pricing CleanMax does not control. Rivalry is intensifying, and the direction of travel is clear — every utility-scale developer with a cost-of-capital advantage now wants the same hyperscaler logos, and ReNew's C&I arm alone had 2.5 GW committed with over 2 GW operational, counting Microsoft, Amazon and Google among its partners.18 Threat of substitutes is real but bounded: a customer's alternatives are the discom (expensive), its own capex-model plant (capital-intensive but increasingly financeable), or eventually batteries and grid-scale storage changing the shape of what round-the-clock clean power costs. Barriers to entry are modest in capital terms and meaningful in capability terms — anyone can build a solar farm; not everyone can structure two hundred group captive arrangements across a dozen state regimes without a compliance failure.
Applying Hamilton Helmer's 7 Powers sharpens the picture further. The two powers CleanMax can credibly claim are process power and counter-positioning. Process power is the accumulated, hard-to-copy operating know-how: fifteen years of opex-model underwriting, a legal and regulatory function that has structured group captive and open-access deals across the country's patchwork of state rules, and — per Antique's diligence — a book where no counterparty has successfully renegotiated a contract against the company since inception.10 Counter-positioning is subtler but real: a utility-scale developer optimised for winning 500 MW auctions with a lean commercial team is genuinely poorly configured to originate and service 1,280 separate PPAs with 588 different corporates. Adapting means building an entirely different sales and credit organisation, which large incumbents are often unwilling to do because it dilutes the metric — gigawatts per employee — on which their existing business is judged.
The powers CleanMax cannot credibly claim are the ones that usually matter most. Scale economies: at 3.5 GW operational, CleanMax is a fraction of Adani Green's 20 GW, and in a business where the primary input cost is capital, scale translates directly into borrowing cost. Network effects: there are none; one customer's solar farm does not make the next customer's more valuable. Switching costs: high during a PPA term, low at its end. Branding: essentially irrelevant when the product is electrons. Cornered resource: no proprietary technology, no exclusive land bank, no unique supply agreement.
So where does CleanMax lose? The most plausible failure mode is not dramatic. It is gradual margin erosion driven by cost of capital. Suppose a large utility IPP with a sovereign-adjacent or investment-grade balance sheet decides that C&I is worth the organisational effort, funds itself two hundred basis points cheaper than CleanMax, and prices accordingly. CleanMax's ₹3.85 per kWh realisation drifts toward ₹3.40. The contracted backlog is protected — existing PPAs are existing PPAs — but every new vintage earns less. Nothing breaks; returns just quietly compress.
The counter-argument, which deserves fair hearing, is that C&I contracting is a service business dressed as an infrastructure business. Customers are not buying a commodity so much as buying certainty: that the plant gets built, that the approvals land, that the savings appear on the electricity bill, that nobody from the state regulator shows up with a surcharge notice. Trust of that kind is earned over years and is genuinely hard to buy with cheap capital. CleanMax's 74% repeat rate and 79% of new capacity from existing customers is the best available evidence that the service layer is doing real work.1011
But note carefully what that evidence does not establish. It shows customers come back during an expansion phase, when the alternative is an unfamiliar vendor and the incumbent is performing. It does not yet show what happens when a well-funded rival offers a 12% lower tariff at renewal. Until a large cohort of PPAs actually reprices in a competitive market, the durability of this advantage remains a hypothesis with supportive but incomplete evidence.
Three consensus claims about this company are worth fact-checking directly, because all three circulate freely and none survives intact.
Claim one: CleanMax is India's largest renewable energy company. It is not, by any measure of capacity. Adani Green operates roughly six times its capacity.1711 CleanMax is the largest specialist supplier to corporate customers, which is a different and much smaller market. The distinction matters because the two businesses compete for capital in the same market and are frequently valued against each other.
Claim two: long-term PPAs make the revenue base effectively locked in. Partially true. Contracts are long, counterparties are strong and no renegotiation has been forced. But contracted capacity is not operational capacity — roughly 2.5 GW of the June 2026 book was still under construction, and construction carries land, grid and supply chain risk that a signed PPA does not eliminate.11 The contracted book is a strong forward indicator, not a bank balance.
Claim three: the data-centre relationship is a moat. The evidence supports scarcity, not lock-in. CleanMax is currently a preferred supplier because bankable clean power in India is short and it has the structuring capability. The same hyperscalers already contract with ReNew's C&I arm.18 Preferred-supplier status held during a shortage is a strong commercial position; it is not the same as a customer who cannot leave.
That distinction — between a story supported by evidence and a story supported by extrapolation — was precisely what India's institutional investors had to adjudicate when the company came to market.
VII. The IPO: A Flop, Then a Reversal
The path to listing began in August 2025, when CleanMax filed a draft red herring prospectus with SEBI for an issue of roughly ₹5,200 crore.2021 By December 2025 the company had filed an updated DRHP, targeting a listing in the third week of that month — a timeline that slipped.22 By the time the offer actually reached investors in February 2026, the size had been cut to ₹3,100 crore: a ₹1,200 crore fresh issue and a ₹1,900 crore offer for sale, priced in a band of ₹1,000 to ₹1,053.1
That trimming — a 40% reduction from the original ambition — was the first signal. Issue sizes come down when bankers test demand and find it thinner than hoped.
The second signal came on February 20, when the anchor book was allotted: ₹921 crore across 87.46 lakh shares at ₹1,053, taken up by 41 institutions.9 The names were as blue-chip as an Indian IPO gets — Temasek, ADIA, HDFC Mutual Fund, Nomura, Franklin Templeton, SBI Life, Premji Invest, Tata Investment Corporation — split roughly 68% domestic and 32% foreign, with nine institutions taking 52.5% of the book.9 On its own, an anchor list like that reads as validation.
Then the book opened to the public on February 23, and the third signal arrived, and it was unambiguous. When subscription closed on February 25, the issue had been covered 0.99 times — technically undersubscribed. The composition told the real story: qualified institutional buyers subscribed 2.99 times their portion, non-institutional investors 0.57 times, and retail investors 0.07 times.1 Retail bid for seven percent of the shares reserved for them.
Read that again, because it is one of the more striking data points in recent Indian primary markets. A retail investor base that has enthusiastically oversubscribed hundreds of small-cap IPOs, that has produced grey-market premiums for companies with a fraction of CleanMax's assets, essentially declined to participate. The proceeds were earmarked largely for debt repayment — roughly ₹1,123 crore of the ₹1,200 crore fresh issue was allocated to repaying borrowings, with the balance for general corporate purposes.9 An IPO whose primary purpose is deleveraging, priced at a premium multiple, in a business retail investors could not easily model, at a moment when the market was cooling on richly-valued new listings.
What followed on March 2 has already been described. The stock opened at a discount, fell as much as 28% intraday, and closed down 17.6% at ₹867.50 on a $341 million raise — the worst first-day showing for an Indian IPO of that size since Paytm.12 It kept sliding to a low of ₹727 by late March.34
At ₹727, CleanMax was valued at roughly ₹8,500 crore. The market was, in effect, pricing a company with 3.1 GW of operating renewable assets, a 5.7 GW contracted backlog, and Brookfield as controlling shareholder, at a level that assumed the growth story would consume more capital than it created.
Then the fundamentals arrived. FY26 results showed revenue up 28% to ₹1,913 crore and net profit multiplying 4.4 times to ₹86 crore.13 The June quarter, reported in late July, showed revenue up 107% year on year and a swing from loss to profit.11 Sell-side coverage initiated: Antique with a Buy in early July, followed by JP Morgan, HSBC and IIFL, all constructive.103 The AI power narrative — which had been in the DRHP all along, just not in a form retail investors could digest — became the market's organising frame.
The stock ran from ₹727 in late March to a peak above ₹1,530 by late June, a gain of more than 110% in roughly three months, before settling back to ₹1,266 by August 12.34
So which market was right? The most defensible answer is that the crash and the rebound were both responses to genuine information, mispriced in magnitude. The March selloff correctly identified that CleanMax carried substantial leverage, generated thin net profit after interest and depreciation, and was priced at a demanding multiple with no operating history as a public company. Those concerns were legitimate and remain legitimate. The June rally correctly identified that the commissioning pace was accelerating faster than consensus assumed, that operating leverage in this model is dramatic once assets are switched on, and that the data-centre pipeline was converting into real megawatts.
What neither move demonstrates is that the market has settled on a correct price. At ₹1,266, CleanMax trades at roughly 95 times trailing earnings on a market capitalisation near ₹14,800 crore.84 Trailing earnings are a poor lens for an infrastructure company mid-build — most of the assets that will generate FY28 profit were commissioned within the last twelve months or are still under construction, and their depreciation and interest are already flowing through the income statement while their revenue is not. That is precisely the argument the bulls make, and it is analytically sound. It is also precisely the argument that gets made about every capital-intensive growth story shortly before the market discovers whether the future earnings actually arrive.
The honest framing: two large price moves in one quarter told us a great deal about sentiment and almost nothing about whether the underlying returns on capital justify either level. For that, one has to go to the statements.
VIII. The Numbers Behind the Story
Start with the top line, because it is the least ambiguous part of this story. Revenue from operations grew from ₹930 crore in FY23 to ₹1,390 crore in FY24, ₹1,496 crore in FY25, and ₹1,913 crore in FY26.8 That is roughly a 27% compound annual rate over the FY23–FY26 stretch, and it accelerated meaningfully in the most recent year. The June 2026 quarter alone delivered ₹832 crore — nearly 44% of the entire prior fiscal year's revenue in a single quarter — up 107% from ₹402 crore a year earlier.11
The profitability arc is equally clear in direction. Net losses of ₹59 crore in FY23 and ₹38 crore in FY24 turned into a ₹19 crore profit in FY25 and ₹86 crore in FY26.8 The June quarter produced ₹55 crore of net profit against a ₹17 crore loss in the prior-year quarter.11
What those numbers describe is a business crossing the threshold where the fixed cost of building an infrastructure platform is finally overtaken by the recurring revenue of running one. The evidence for that interpretation is in the overhead line: selling, general and administrative expenses fell to 8.7% of RE Power Sales income in the June quarter, from 18.2% in FY23.11 The company roughly doubled its revenue while its corporate cost base grew far more slowly. That is genuine operating leverage, and it is the single most encouraging structural fact in the financials.
Now turn the page, because the same statements contain a harder story.
Operating margin in the June 2026 quarter came in at 50.6%, down from 66.1% a year earlier — a compression of more than fifteen percentage points.23 Management and sympathetic coverage attributed this to rising employee and operating costs as the business scales. That explanation does not survive contact with the detail: employee costs in the quarter were ₹32.27 crore, versus ₹31.64 crore a year earlier — an increase of about 2%.23 Employee costs did not cause a 1,553 basis point margin decline.
More importantly, this was not a one-quarter aberration. Operating margin was 63.2% in the September 2025 quarter, 62.3% in December 2025, then fell to 48.0% in the March 2026 quarter before the 50.6% print in June.2423 Two consecutive quarters near or below 50%, following two quarters in the low sixties, is a trend, not noise.
The most likely explanation is mix, and it is not necessarily sinister. As CleanMax's RE Services segment executes larger EPC contracts, and as newly commissioned open-access and grid-connected projects come online with different cost structures than legacy rooftop assets, blended margin falls even if each individual business is performing as designed. An EPC rupee simply carries less margin than a power-sales rupee. But the company has not, in its public materials, decomposed the compression into mix versus pricing versus cost. That is a disclosure gap, and it is the first thing an analyst should press on the next earnings call. Until it is answered, an investor cannot distinguish "we sold more low-margin EPC" from "we are winning power sales contracts at worse tariffs."
One caution about the June quarter specifically: it flatters the business seasonally. India's south-west monsoon season is peak wind generation, and the pre-monsoon months deliver peak solar irradiance. A renewables portfolio weighted toward wind and hybrid assets therefore earns disproportionately in the April-to-September half, and disproportionately less in the winter quarters. The December 2025 quarter's revenue of ₹422 crore against ₹557 crore in March 2026 and ₹832 crore in June 2026 reflects that rhythm as much as it reflects growth.2411 Annualising a June quarter in this business overstates the run rate. Year-on-year comparison is the only honest one.
There is a related earnings-quality question. In the March 2026 quarter, other income of ₹82 crore represented roughly 59% of profit before tax.24 Other income in an infrastructure holding company typically comprises treasury income, fair-value adjustments and one-off items. When it accounts for the majority of pre-tax profit in a quarter, the headline profit number is telling you less about the operating business than the number suggests. This does not make the profit fake. It makes it less repeatable, and investors should weight the operating line accordingly.
Then there is the balance sheet, which is where the real debate sits.
Total borrowings grew from ₹1,341 crore in FY23 to ₹12,684 crore in FY26 — close to a tenfold increase in three years.8 Long-term debt alone rose from ₹7,127 crore in March 2025 to ₹11,312 crore in March 2026, a 59% jump in twelve months.23 Interest expense in the June quarter reached ₹254.71 crore, the highest quarterly figure in the company's history and 15.5% above the prior year.23 Total assets grew from ₹6,873 crore to ₹22,555 crore over the same three-year span, and fixed assets stood at ₹11,954 crore in March 2026, up 48% year on year.823
Set against those figures, the coverage ratios look tighter than the growth optics suggest. Debt to EBITDA stood at 1.96 times and interest coverage at 2.54 times.23 Neither is alarming for contracted infrastructure. Both are numbers that leave less room than an equity investor accustomed to asset-light businesses might assume — and Antique's own model projects net debt to EBITDA reaching roughly 5 times by FY28 on approximately ₹21,900 crore of debt as the build-out continues.10 That is the bull case's own leverage forecast, not the bear case's.
Two further items deserve attention. Trade payables rose from ₹1,295 crore in March 2025 to ₹3,481 crore in March 2026 — a 169% increase against 48% growth in fixed assets.23 For a company in a heavy construction cycle, stretching supplier payments is normal working capital management and partly reflects equipment purchases in flight. It is also a form of unrecognised financing, and if it continues to outgrow the asset base it becomes a question worth asking. And cash flow: FY26 operating cash flow was ₹1,731 crore, investing outflow was ₹5,953 crore, and free cash flow was negative ₹4,039 crore.8 Return on capital employed was 6%.8
That last pair of numbers is the fulcrum of the entire investment case, so let it stand without decoration. CleanMax generates substantial operating cash and spends far more than it generates building new assets. It is financing the gap with debt and equity. Whether that is value creation or value destruction depends entirely on whether the return on each incremental gigawatt exceeds the blended cost of the capital funding it.
Here the news is genuinely encouraging on one side of that equation. The weighted average cost of project debt fell to 8.4% in the June quarter from 9.2% in April 2025, and the company carries a CARE AA- (Stable) rating.11 An 80 basis point reduction across a ₹12,000 crore debt stack is worth close to ₹100 crore annually at run-rate — real money, and evidence that scale and the Brookfield association are translating into cheaper funding. The company also moved in July 2026 to consolidate 148 MWp of rooftop SPVs into the holding company and to approve a domestic bond issuance, both aimed at diversifying funding sources and locking in fixed rates.11 These are the actions of a finance team managing the liability side deliberately rather than passively.
What remains unproven is the asset side. Nobody outside the company can currently compute the return on a specific project vintage, because CleanMax does not disclose per-project or per-vintage economics. Management's guidance of a minimum ₹3,000 crore of EBITDA by FY28 implies a substantial step-up from FY26's ₹1,132 crore operating profit.38 If that materialises on the debt base Antique projects, the returns work. If margin compression continues while leverage climbs, they do not. Both outcomes are consistent with everything currently disclosed — which is exactly why the stock could trade at ₹727 and ₹1,533 in the same quarter without either price being obviously irrational.
The people making those capital allocation decisions therefore matter more than usual.
IX. Management Credibility & Capital Allocation
There is a particular kind of founder produced by elite consulting: analytically rigorous, fluent in capital structure, comfortable presenting to institutional investors, and — the usual criticism — sometimes better at designing businesses than operating them. Kuldeep Jain has now spent fifteen years disproving the second half of that stereotype in one of the least forgiving operating environments imaginable, which is the Indian power sector.
His track record contains an unusually clean early failure and a clear recovery from it. Betting the company on gas-fired generation in 2010 and watching Fukushima destroy the economics within a year is the sort of thing founders usually launder out of the official narrative. Jain has discussed it openly, including the personal financial strain.5 That candour is a small signal, but a real one: managers who describe their failures accurately tend to describe their present accurately too.
The more testable question is target-setting. In June 2023, alongside the Brookfield transaction, Jain publicly committed to becoming a 5 GW-plus platform in three to four years and adding 800–1,000 MW annually.12 By FY26 the company commissioned roughly 1.4 GW in a single year and reached 5.7 GW contracted, with 6.8 GW by the June 2026 quarter.1311 Set a specific multi-year target, exceed it. In a sector where Indian developers routinely announce pipelines that never get built, that is a meaningful credential.
At IPO, Jain sold shares in the offer for sale — reported at approximately ₹321 crore — while retaining a substantial stake. Reasonable people can disagree about founder selling at listing; after fifteen years of illiquidity it is normal and arguably healthy for the founder to diversify. What matters is what remained, and Jain remains the largest individual non-institutional shareholder within a promoter group holding 49.4%.8 His personal wealth is still overwhelmingly tied to the equity, which is the alignment that counts.
Nikunj Ghodawat, the chief financial officer, has been the more visible voice on the financial narrative since listing. On the Q1 FY27 call held on August 3, 2026, his framing emphasised "improving scale, lower borrowing costs, and disciplined financial management" as the supports for the expansion pipeline, while Jain reiterated that the company remained on track for its annual growth targets.11 The language is consistent with what management said pre-IPO and at FY26 results — no narrative shift, no quiet abandonment of a metric.
But consistency is not the same as completeness, and here is where a skeptical investor should push. Management's public commentary since listing has led with revenue growth, adjusted EBITDA growth, capacity additions and cost of debt. It has not, in the materials available, directly addressed why operating margin fell from the low sixties to around 50% across two consecutive quarters. Leading with adjusted EBITDA growth of 74% while reported operating margin compresses by fifteen points is technically accurate and analytically incomplete. It is not deception; it is emphasis. And emphasis is exactly what an investor should track over the next several calls, because the difference between a management team that pre-empts an uncomfortable trend and one that waits to be asked is one of the more reliable tells in the business.
On capital allocation, the record to date is refreshingly simple: CleanMax has built rather than bought. There is no history of acquisitions, no adjacent-industry diversification, no vanity vertical integration into module manufacturing — which several Indian renewables companies have attempted with mixed results. Every rupee has gone into contracted generation capacity in the business it understands. In a market where "diworsification" has destroyed a great deal of Indian infrastructure shareholder value, that discipline deserves explicit credit.
Two governance items belong in an honest assessment. At the 16th annual general meeting on July 24, 2026 — the first as a listed company — shareholders approved 56 resolutions, of which 43 were material related party transactions involving subsidiaries and joint ventures including Clean Max Ajanta, Clean Max Terra, Clean Max Vayu and Kanoo Cleanmax Renewables, alongside corporate guarantees of ₹474.27 crore for subsidiaries.25 Forty-three related party resolutions sounds alarming until one understands the structure: every project sits in its own special purpose vehicle for financing reasons, and transactions between the parent and its own SPVs are technically related party transactions. Promoters abstained as required, and approval margins ran between 93.2% and 99.99%.25 This is structural complexity, not a red flag — but it is complexity, and complexity is where problems hide. An investor should watch whether SPV-level disclosure keeps pace with SPV-level proliferation. The July 2026 decision to consolidate 148 MWp of rooftop SPVs into the holding company moves in the right direction.11
The second item: the re-appointment of director Murzash Manekshana passed with 98.6% support but drew 6.33% dissent specifically from public institutional shareholders.25 That is a modest signal in isolation. It is worth noting only because institutional dissent on board composition, in a company where a private-market controlling shareholder appoints much of the board, is the mechanism through which minority concerns first surface.
The candid limitation on all of this: CleanMax listed in March 2026. There is no multi-year public guidance track record, no observed instance of the company missing a number and explaining why, no test of how management behaves under public-market pressure. The pre-IPO execution record is good. The post-IPO credibility record does not yet exist.
Which means the case has to be argued on structure and evidence rather than on trust.
X. Bull Case vs. Bear Case
The bull case, stated at its strongest.
India's electricity demand is growing structurally, its industrial base is expanding, and its corporate sector faces mounting pressure — from global customers, from supply chain requirements, from its own commitments — to decarbonise. CleanMax sits at the intersection of those forces with a fifteen-year head start in the specific niche where the economics are best.
The forward visibility is unusually concrete for a growth company. A 6.8 GW contracted portfolio against 3.5 GW operational means roughly half the future revenue base is already under signed contract, not forecast.11 These are fifteen-to-twenty-five year agreements with counterparties of which more than 95% are rated A- or better, and per Antique's diligence, not one has been renegotiated against the company since inception.10 Contracted, credit-worthy, long-duration cash flow is the most valuable characteristic an infrastructure business can have.
The data-centre exposure is not a speculative narrative overlay; it is 2.5 GW of contracted capacity with named counterparties.11 Apple has moved from customer to co-investor.15 The pricing advantage over utility-scale — roughly ₹3.85 versus ₹2.50 per kWh of EBITDA realisation — is structural, arising from the fact that Indian industrial grid tariffs are inflated by cross-subsidy, which sets a high reference price for bilateral contracts.10
Operating leverage is demonstrable: overhead at 8.7% of power sales income against 18.2% three years ago, and cost of debt down 80 basis points in fifteen months.11 Brookfield's control means capital access is not the binding constraint it was in 2022. Antique models a 48% revenue CAGR and 63% EBITDA CAGR from FY26 to FY28, with capacity reaching 7.8 GW by FY29.10 Management has guided to at least ₹3,000 crore of EBITDA by FY28.3
The bear case, stated at its strongest.
Begin with what the primary market told us six months ago. This company came to market with Temasek, ADIA and a full complement of blue-chip anchors, and could not get its book covered. Retail subscribed 0.07 times.1 The size had already been cut 40% from the original filing.201 The proceeds were overwhelmingly for debt repayment.9 The market's first collective judgement on this asset, made with the full prospectus in hand, was a shrug — and the price fell 17.6% on day one.2 The subsequent rally happened on two quarters of results and four sell-side initiations. Which of those two assessments involved more diligence is a fair question.
Valuation leaves little margin for error. At roughly 95 times trailing earnings, the price embeds substantial future execution.8 Yes, trailing earnings understate an infrastructure company mid-build — but that argument requires the future earnings to arrive on schedule, and the schedule requires roughly doubling operating capacity by FY29 while margins are compressing and debt is climbing.
The margin trend is the most concrete bear evidence available. Two consecutive quarters near or below 50% operating margin against low-sixties a year earlier is not a rounding error, and the company's own explanation — rising employee and operating costs — is contradicted by a 2% increase in employee costs.2324 Something else is driving it, and until management decomposes it, the possibility that new PPA vintages carry structurally worse economics cannot be excluded.
Leverage compounds the sensitivity. Borrowings grew roughly tenfold in three years to ₹12,684 crore, quarterly interest expense hit a record ₹254.71 crore, free cash flow was negative ₹4,039 crore in FY26, and ROCE was 6%.823 The bull case's own analyst projects net debt to EBITDA around 5 times by FY28.10 In a business where the primary input is borrowed money, a sustained rise in Indian rates or a tightening of credit conditions transmits directly to project returns and to the value of the equity stub sitting above ₹22,000 crore of assets.
Regulation is the wildcard nobody controls. The Electricity (Amendment) Rules, 2026, notified on March 13, 2026 and largely effective April 1, rewrote group captive rules — broadly helpfully, by allowing corporate groups to aggregate holdings toward the 26% ownership threshold — but also introduced a proportionality cap under which consumption beyond a user's ownership share attracts cross-subsidy and additional surcharges.7 More consequentially, the full waiver of interstate transmission charges for renewables ended June 30, 2025; projects commissioned after that face charges rising 25% annually until the full 100% applies from July 2028.26 The waiver was worth roughly ₹0.70 per kWh.26 Every new interstate open-access project CleanMax builds from here carries a progressively higher landed cost than the projects in its existing portfolio — a headwind to new-vintage economics that is scheduled, known, and does not go away.
Competitive intensity is the slow risk. ReNew's C&I arm already has 2.5 GW committed with Microsoft, Amazon and Google.18 Tata Power claims the top position in Indian rooftop solar, having installed 1 GWp in nine months of FY26.19 Adani Green's cost of capital at 20 GW of scale is not CleanMax's cost of capital at 3.5 GW.17
An activist's stress test. A skeptical long-short investor looking at this business would push on four specific things. First, disclosure: the company reports adjusted EBITDA prominently and does not break out per-vintage project returns, making it impossible to verify whether incremental capital is earning above its cost. Second, earnings quality: other income at 59% of pre-tax profit in a quarter deserves explanation, not a footnote.24 Third, structural complexity: dozens of SPVs, 43 material related party resolutions and ₹474 crore of corporate guarantees create a consolidation picture that is difficult to audit from outside.25 Fourth, and most pointedly: with Brookfield holding effective control through a fund with a finite life, whose capital allocation preferences govern when growth and exit timing conflict? None of these are accusations. All four are questions a serious investor should be able to answer before underwriting the equity, and only the first is likely to be answered voluntarily.
The synthesis is that the bull and bear cases are not arguing about different facts. They are arguing about a single question: whether the return on each new gigawatt is holding up. Everything else follows from that.
XI. Durable Business & Investing Lessons
The first lesson is about the nature of innovation, and it runs against the instinct of most growth investors.
CleanMax invented nothing. It does not manufacture solar modules — those come from a brutally commoditised global supply chain. It does not make turbines. It holds no patents that matter. Every physical component of a CleanMax project can be bought by anyone with a purchase order.
What CleanMax built was a financing structure. The opex model took a capital expenditure decision that Indian corporate CFOs would reliably defer and converted it into an operating expense decision that a plant manager could approve on cost savings alone. That single reframing — the same technology, the same electricity, a different balance sheet — created a category leader.
This pattern recurs across business history far more often than technological breakthrough does. Aircraft leasing did not change aviation technology; it changed who owned the asset, and it transformed the industry. Software-as-a-service did not change what enterprise software did; it changed how it was paid for, and it rebuilt the entire sector's economics. The generalisable insight for investors is that when a product's economics are compelling but adoption is blocked by who bears the capital and the risk, the company that solves the ownership question often captures more durable value than the company that improved the product.
The corollary, which CleanMax's financials illustrate uncomfortably well, is that whoever takes the balance sheet risk must be paid for taking it — and must be scrutinised for whether they are. A 59% operating margin that converts to 6% return on capital employed is not a contradiction; it is the arithmetic of an asset-heavy model. Investors who see the margin and infer a high-return business have made a category error.
The second lesson concerns institutional capital, and it is genuinely two-sided.
Brookfield's investment did something no amount of operational excellence could have accomplished: it removed the capital ceiling. A company that added a few hundred megawatts a year became one that added 1.4 GW a year. That is not incrementally better; it is a different company competing in a different weight class, and it happened because of who owned it rather than what it did differently.
But permanent-seeming capital comes attached to a clock. Private infrastructure funds have defined lives and return obligations to their own investors. A controlling shareholder whose fund is early in its life is the best possible partner for a growth-stage public company; the same shareholder late in a fund's life may prioritise realisation. Public shareholders in this structure hold a genuinely good asset alongside a genuinely good investor whose exit timing they do not control and cannot easily forecast. The lesson is not to avoid such structures. It is to price the fact that you are not the one deciding when the story ends.
The third lesson is about how markets process complex, fast-scaling infrastructure — and it is the most immediately useful.
Within a single quarter, the same public market valued this company at ₹8,500 crore and roughly ₹18,000 crore. No new plant was invented. No contract was won that had not already been signed. What changed was the availability of a simple story. In March, the pitch was "leveraged solar developer with thin profits at a demanding multiple," which is accurate. By June, the pitch was "AI power play with 42% data-centre exposure and 107% revenue growth," which is also accurate. Both descriptions fit the identical set of facts.
Markets are efficient at processing simple narratives quickly and inefficient at processing complex ones at all. A business whose value depends on the incremental return on capital across dozens of project vintages, under state-by-state regulation, funded with layered project debt, is genuinely hard to model. When something is hard to model, the market substitutes a story — and stories are volatile in ways that unit economics are not. The investor's edge, if one exists here, comes from doing the modelling that the narrative substitutes for.
Which brings the analysis to the specific things that could break, and the specific things worth counting.
XII. Risk Radar
Regulatory and state policy risk sits at the top, because it is the only risk here that can reprice a contract already signed. India's electricity regulation is not one regime but roughly thirty. Each state regulator sets its own cross-subsidy surcharge, additional surcharge, banking charges — the terms on which a generator can "store" surplus generation with the grid and withdraw it later — and open-access approval process. These are periodically revised, and revisions have historically gone in the direction that protects discom revenue, because discoms are financially stressed and every megawatt a factory buys from CleanMax is a megawatt of high-margin industrial load the discom loses. A state that materially raises banking charges or narrows open-access eligibility can compress the realised economics of projects already operating in that state.
The specific, dated version of this risk is the ISTS transmission charge phase-out. Full waiver applied to renewables commissioned through June 30, 2025; a 25% levy applies to projects commissioned between July 2025 and June 2026, escalating 25% annually until 100% of charges apply from July 2028.26 With the waiver worth roughly ₹0.70 per kWh, the cost base of new interstate open-access projects rises on a known schedule.26 Either CleanMax passes that through — narrowing its discount to grid tariffs, which is the whole value proposition — or it absorbs it in margin. This is not a tail risk. It is a scheduled headwind already partly visible in the compression the numbers have shown.
The group captive rules cut both ways. The March 2026 amendments broadened ownership recognition across corporate groups, which genuinely helps CleanMax structure deals for large conglomerates, but introduced a proportionality cap under which consumption above a user's ownership share attracts the surcharges the structure exists to avoid.7 Verification also moved from the Central Electricity Authority to the National Load Despatch Centre for interstate projects, with a new grievance mechanism.7 The net effect is probably modestly positive; the important point is that the rules governing the core structure of CleanMax's business changed twice in recent years and will change again.
Refinancing and rate risk is the mechanical one. With ₹12,684 crore of borrowings and record quarterly interest expense, the equity value is a residual claim on a large, leveraged asset base.823 Project debt in Indian renewables typically carries reset clauses; a sustained upward move in benchmark rates flows through to a substantial portion of the stack. The cost of debt falling to 8.4% is genuine progress and reflects both scale and sponsor quality, but it also means the current earnings trajectory partly depends on funding conditions staying benign.11 The domestic bond issuance approved in July 2026 is a sensible attempt to term out and fix that exposure.11
Customer concentration is the debated one. Data centres and AI represent 42% of contracted power sales capacity, up from a negligible base two years ago.11 The counterparties are among the world's most creditworthy, which sharply limits credit risk. The exposure is to capex plans, not solvency. Global AI infrastructure spending has been revised in both directions before, and a contracted megawatt not yet under construction is a softer commitment than an operating asset. The more probable manifestation, as noted earlier, is pricing pressure from sophisticated repeat buyers rather than contract loss.
Execution risk is arithmetic. Reaching 7.8 GW by FY29 from 3.5 GW operational requires sustained commissioning at recent record rates for three consecutive years.1011 Each gigawatt needs land aggregation, grid connectivity, state approvals, module and turbine supply, and financing — simultaneously and repeatedly. CleanMax has executed well recently, including a record 0.5 GW quarter and a 525 MWp central-transmission-connected project. But the FY29 target requires no sustained failure in any of those inputs. Grid connectivity in particular is a constraint the company does not control; India's transmission build-out has lagged its generation build-out for years.
Competitive intensity is the slow bleed. The mechanism is not dramatic loss of business but progressive tariff compression as better-capitalised utility IPPs pursue the same corporate customers with cheaper funding. Because PPAs are long-dated, this would appear gradually — in the returns of new vintages, invisible in headline growth for years.
Two second-layer items round this out. The CARE AA- (Stable) rating is a meaningful credential for a recently listed, rapidly levering issuer, and any change in outlook would be an early warning worth watching.11 And the concentration of assets in dozens of project SPVs, with 43 material related party resolutions approved at the first post-listing AGM, means consolidated statements aggregate a structure that is complex by design.25 Complexity is not wrongdoing. It does mean disclosure quality matters more here than in a simpler business.
Given all of that, most of what an investor needs to monitor collapses into a very short list.
XIII. KPIs That Matter Most
Three metrics, and only three, carry most of the information about whether this business is working.
First: the conversion of contracted capacity into operational capacity. As of June 30, 2026, CleanMax reported 3.5 GW operational against 6.0 GW of contracted RE Power Sales capacity, with 2.5 GW under execution.11 The gap between those numbers is the entire growth story, and closing it is the entire execution question. A contracted megawatt earns nothing. Every quarter, the investor should ask a simple question: how many megawatts moved from "under execution" to "operational," and is the pace consistent with management's guidance of at least 1.5 GW of additions in FY27 and roughly 7.8 GW of capacity by FY29?1110 Commissioning pace is the cleanest, least manipulable read on whether land, grid connectivity, supply chain and financing are all working at once. It is also a leading indicator — a slowdown in commissioning shows up long before it shows up in revenue.
Second: EBITDA margin trajectory, watched at the reported operating level rather than the adjusted level. The compression from roughly 63% to around 50% across recent quarters is the single most important unresolved question in the financials.2423 The investor should track reported operating margin quarter by quarter, and — more importantly — listen for whether management decomposes it. A specific explanation ("this quarter's mix included X crore of lower-margin EPC revenue; power sales margins were stable at Y%") would substantially resolve the bear case. Continued attribution to generic "scaling costs," against employee expenses that grew 2%, would not. If margins stabilise in the low fifties on a clear mix explanation while capacity doubles, the model works. If they keep drifting down, growth is being purchased with returns, and the valuation logic collapses.
Third: leverage and coverage, taken together. Debt to EBITDA at 1.96 times and interest coverage at 2.54 times are the current readings.23 The bull-case model itself projects net debt to EBITDA rising toward 5 times by FY28 as the build-out proceeds.10 Both numbers should be watched against commissioning progress, because leverage rising ahead of capacity is the specific pattern that precedes trouble in infrastructure businesses: debt is drawn when construction starts, EBITDA arrives when the plant switches on, and the gap between those two events is where balance sheets get into difficulty. Rising leverage with rising operational capacity is a build-out. Rising leverage with flat operational capacity is a warning.
Everything else — customer count, contracted portfolio milestones, individual marquee logos — is supporting detail. These three tell the story.
A note on what deliberately is not on this list. Contracted portfolio growth is the metric the company leads with, and it is genuinely informative, but it is also the easiest to grow: signing a PPA costs a sales cycle, while delivering one costs several hundred crore of capital. A company can grow its contracted book indefinitely while its commissioning pace stalls, and the reported number would look excellent throughout. Customer count has the same weakness — 593 logos says nothing about the megawatts or the tariffs behind them. Adjusted EBITDA is excluded for the reason already given: it is management-defined, and the gap between it and reported operating profit is precisely where the current uncertainty lives.
XIV. Epilogue: What to Watch Next
The next four quarters will resolve most of what is currently unknowable about CleanMax.
The nearest test is the September 2026 quarter. If reported operating margin stabilises and management explains the compression with specifics rather than macro language, the most credible bear argument weakens considerably. If margins slip again and the explanation stays generic, an investor should assume the new-vintage economics are genuinely worse and reprice accordingly. Beyond margin, the number to watch is commissioning: management has guided to more than 1.5 GW of additions across the remainder of FY27, which requires sustaining something close to the record pace set in the June quarter.11
The financing side will also reveal something. The domestic bond issuance approved in July 2026 is the company's first significant move to diversify beyond project debt as a listed entity.11 The pricing it achieves is a market verdict on the credit that is independent of the equity narrative — bond investors and equity investors look at the same balance sheet with very different priorities, and when they disagree, the bond market has historically been the better guide. Whether the build-out requires an additional equity raise before FY29 is the other open question; funding roughly ₹22,000 crore of projected FY28 debt against a growing but still modest equity base leaves that possibility live.10
International expansion remains a footnote, and should be treated as one. CleanMax operates in the Middle East, with more than 80 MWp of solar assets across the region and a Dubai presence, and entered Thailand with support from the International Finance Corporation for greenfield and refinanced C&I solar capacity. Together these accounted for under 6% of FY25 revenue. They are genuine optionality — the C&I opex model travels reasonably well to markets with high industrial tariffs and rising corporate decarbonisation pressure — but they are not part of the current investment case, and any framing that leads with international growth is leading with the wrong thing.
The larger frame is this. CleanMax's core story is not manufactured. India genuinely needs enormous quantities of clean industrial power. Global technology companies genuinely need bankable renewable supply in India and have limited alternatives. CleanMax genuinely has fifteen years of accumulated capability in the specific structuring problem those needs create, a contracted book approaching 7 GW, a controlling shareholder with deep pockets, and an operating record over the past three years that met the targets it set publicly.
What remains unproven is the part that determines whether any of that translates into shareholder returns: whether the return on each incremental gigawatt exceeds the cost of the capital that funds it, in a business where margins have compressed for two consecutive quarters, borrowings have grown roughly tenfold in three years, and free cash flow is deeply negative by design.
The market has already demonstrated, twice within a single quarter, how far sentiment can travel while that question stays open. In March it priced the leverage and ignored the backlog. In June it priced the backlog and ignored the leverage. Neither move was informed by new evidence on the only thing that ultimately matters. That evidence arrives one quarterly commissioning report and one margin line at a time — and for a company barely six months into its public life, the record that will settle the argument has scarcely begun to accumulate.
References
-
Clean Max Enviro Energy Solutions IPO — issue structure, subscription, listing price and holdings — IPOJI ↩↩↩↩↩↩↩
-
Clean Max Shares Tumble in Mumbai Debut After $341 Million IPO — Bloomberg, 2026-03-02 ↩↩↩↩
-
After muted listing, this stock turns multibagger in just 3 months; should you still buy? — Business Today, 2026-08-05 ↩↩↩↩↩↩
-
CLEAN MAX ENVIRO EN SOL L (CLEANMAX.NS) — NSE Company Quote Page ↩↩↩↩
-
How Kuldeep Jain's CleanMax is greening India Inc — Forbes India ↩↩↩↩↩↩↩↩↩
-
SEBI — Clean Max Enviro Energy Solutions Limited DRHP Filing, 2025-08 ↩↩
-
Reforming Captive Power in India: Key Changes under the Electricity (Amendment) Rules, 2026 — pv magazine India, 2026-03-31 ↩↩↩↩
-
Clean Max Enviro Energy Solutions Ltd — Screener.in Consolidated Financials ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩
-
Clean Max Enviro Energy raises INR 921 crore from anchor investors (anchor book, pre-IPO round and use of proceeds) — pv magazine India, 2026-02-23 ↩↩↩↩↩↩
-
Stock to buy: Clean Max shares offer 31% upside, says Antique — Business Today, 2026-07-02 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩
-
CleanMax Reports Strong Q1 FY27 Results With Revenue Doubling And Renewable Energy Capacity Growth — SolarQuarter, 2026-08-01 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩
-
CleanMax raises $360 million from Brookfield — pv magazine India, 2023-06-02 ↩↩↩↩
-
CleanMax Reports 28 Percent YoY Revenue Growth in FY26 to INR 1,913 Crore, PAT Jumps 4.4x — Energetica India ↩↩↩↩↩
-
India's data centre capacity additions grow 59 per cent in H1 2026, says industry report — Indian Infrastructure, 2026-07-22 ↩
-
Apple to Invest ₹1 Billion in CleanMax for 150 MW Renewable Energy Projects — Mercom India, 2026-05-08 ↩↩
-
Apple partners CleanMax to build 150 MW of renewable energy capacity for captive consumption in India — pv magazine India, 2026-05-07 ↩
-
Adani Green Energy surpasses 20 GW operational capacity — Adani Group Media Release, 2026-07-01 ↩↩↩↩
-
ReNew Commissions 2.4 GW of Renewable Energy Capacity in FY2026 — ReNew Press Release, 2026-04-15 ↩↩↩↩
-
Tata Power Renewables Achieves 1 GWp Rooftop Solar Installation Capacity in 9 Months (FY26) — Tata Power Media Release ↩↩
-
CleanMax to raise ₹5,200 cr via IPO; DRHP filed with Sebi for approval — Business Standard, 2025-08-18 ↩↩
-
Clean Max Enviro IPO: Profit swings, cost overruns; 5 key risks to watch — Business Standard, 2026-02-19 ↩
-
Clean Max Enviro files updated DRHP with Sebi; IPO planned for 3rd week of December — Business Today, 2025-12-07 ↩
-
Clean Max Enviro Q1 FY27: Stellar Revenue Growth Masks Margin Compression Concerns — MarketsMojo ↩↩↩↩↩↩↩↩↩↩↩↩↩
-
Clean Max Enviro Energy Solutions Q4 FY26: Strong Profit Surge Masks Margin Compression Concerns — MarketsMojo ↩↩↩↩↩↩
-
Clean Max Enviro shareholders approve 56 resolutions at 16th AGM — ScanX, 2026-07-24 ↩↩↩↩↩
-
Green Energy Open Access Developers Want ISTS Charges Waived Beyond June 2025 — Mercom India ↩↩↩↩
LinkedIn
WhatsApp
Reddit
Hacker News
Email
Kindle
RSS Feed
Spotify
Apple Podcasts
Amazon Music
Audible
YouTube