77.4% ROCE, 26.2% operating margin, zero debt — the highest return anywhere on this map, earned on almost no capital employed. The moat is certification and lead time, not technology: 765 kV transformers, GIS and HVDC gear are supplied in India by roughly three credible names, qualification with the transmission utilities takes years, and global transformer lead times are running in years. Data-centre load and the national grid rebuild pull the same product, so it is levered to two independent demand cycles. Sustained sales growth of 12.4% understates it — profit compounded 85.6% on the lower of the 3- and 5-year windows. Against it: 82x earnings already capitalises the order book, and this is an order-book business, not an annuity.
Best positioned in each value chain
The maps show where every company sits. This page asks the next question: within each chain, which three or four companies are actually best placed — on growth, on margin, on return on capital, and on how defensible the position is. Eighty-six companies across twenty-two of the site’s twenty-six chains, screened from 890 listed tickers and then argued one by one, with the case against each stated alongside the case for it.
How these were chosen
- Sustained growth is min(3-year, 5-year sales CAGR), not the 5-year figure — the same rule the leader strip on every map uses. A company has to have compounded over both windows, which removes base effects without a hand-maintained exclusion list.
- Returns are ROCE for operating companies and ROE for lenders and fee businesses, where ROCE carries no information.
- Margin is operating margin scored against the median of that chain, not against a fixed bar — a 3.6% margin in contract assembly and a 56% margin in mining are both good, in context.
- Screening floors: ₹2,000 Cr market cap and ₹500 Cr revenue. The composite ranks growth, return, margin and profit CAGR within each chain, with a penalty for debt-to-equity above 2.0 outside financials.
- The moat argument is judgement, not screen output. The screen produces a shortlist; the reasoning below states what the barrier actually is and what would break it. Where the top-scoring name is a base effect, a spin-off artefact or a subsidy programme, it is named and excluded in writing rather than quietly dropped.
- Every figure is generated from the data, not transcribed — 302 numeric claims in the reasoning were checked against the Screener export programmatically.
The best economics in India's AI chain are not in AI. They are in the electrical equipment that energises a data centre: two suppliers on this map earn 68–77% on capital with no debt, against 13% for the largest domestic AI-server maker. Compute is assembled here; the grid hardware is gated.
67.8% ROCE, 21.4% margin, zero debt — the second-best return on the map, on the same structural position as GE Vernova. Against it: demerged too recently to have a 3- or 5-year growth record, so the sustained-growth screen cannot rank it at all. You are buying the franchise on one year of accounts.
Sustained sales growth of 69.9% — the fastest on the map — at 37.5% ROCE. The only listed Indian-origin AI server and HPC maker with its own design (Tyrone), which matters because government and PSU procurement carries a local-content preference that a rebadged import cannot claim. Against it: a 13.3% operating margin says this is high-value integration, not silicon; the moat is a procurement rule, which can be rewritten. At 113x, nothing can go wrong.
84.9% operating margin and 51.8% ROCE at 23x — a marketplace that owns no assets and takes no position, in the layer where data centres will contract flexible power. Against it: market coupling is the single biggest threat on this map — it is designed to remove exactly the network effect that produces this margin. 13.9% sustained growth is unremarkable. This is a cheap option on a business with a named, dated regulatory risk, not a compounder.
Excluded: NTPC Green Energy screens first on growth (156%) but earns 3.6% on capital at 128x — a pipeline being commissioned, priced as if it were already earning.
The highest returns on equity in the auto chain belong to lenders, not manufacturers, and the best manufacturer return belongs to the one with brand pricing power rather than scale. Component suppliers earn their way up only where an aftermarket sits alongside the OE contract.
30.5% ROCE, 24.7% operating margin, zero debt — the only Indian OEM with genuine pricing power, because Royal Enfield does not compete on price in the 250–650cc leisure segment it effectively defined and still dominates. Margin at nearly 25% against a mid-teens industry is the moat made visible. VECV adds commercial-vehicle scale without diluting the return. Sustained growth 17.5%. Against it: 38x, and the export/EV challenge to the mid-size motorcycle franchise is real but slow.
36.0% ROCE, 16.1% margin, zero debt, 22x — the highest return among OEMs on this map, from vans, LCVs and contract engine assembly for Mercedes-Benz and BMW in India. Contract engine work is qualification-gated and multi-year: the customer cannot switch quickly. Profit compounded 65.5%. Against it: a narrow product set and a small number of customers; one lost contract is material.
26.6% sustained sales growth, 19.4% ROE — the fastest-growing large vehicle financier. Used commercial-vehicle lending is the highest-yield book in Indian vehicle finance and the hardest to underwrite; two decades of repossession and resale data in the Murugappa franchise is a data moat that a new entrant cannot buy. Against it: 6.9x leverage. This is a credit-cycle business and freight rates are the cycle.
34.5% sustained growth, 26.4% ROE, 21.2% ROCE — a Tier 1 supplier whose aftermarket distribution arm is what lifts the return above a pure OE captive. OE supply alone is a margin-squeezed business; the same parts sold into replacement demand are not. Against it: ₹13,787 Cr market cap, D/E 1.0, and OE customer concentration remains.
Excluded: Bajaj Finance is the larger and better-known franchise but grows slower (25.2%) at 3.8x leverage and 33x; Cholamandalam is the same trade with more growth.
Manufacturing crop chemicals is not where the money is. The margin sits with whoever owns the registration or the molecule — and, further along, with whoever owns the farmer's wallet. Fertiliser manufacture is a subsidy pass-through and prices like one.
29.1% ROCE, zero debt, 25x — patented molecules plus a seed-and-chemistry bundle sold through a rural brand built over decades. The moat is the patent and the brand together: neither alone survives generic entry. Against it: 3.4% sustained growth and a flat quarter. You are buying quality and a 3.6% yield, not a growth rate.
30.2% ROCE, zero debt, at 11x earnings — the highest return in the chain, from a model that owns roughly 2,900 product registrations in Europe and Latin America and outsources every gram of manufacturing. The registration is the asset: it takes years and money to obtain and cannot be replicated by adding a plant. Against it: 9.2% sustained growth, and generic post-patent pricing is where its revenue lives.
33.9% operating margin, 21.0% ROCE, 16.8% sustained growth, TTM sales +42.4%. Fluorination is the barrier — handling anhydrous HF safely at scale is a capability, not a capex line, and a rising share of new crop-protection actives are fluorine-bearing. That makes this a structural share gain rather than a cyclical one. Against it: 53x prices the structural argument in full.
20.3% ROE, 17.9% sustained growth, 22x — India's largest tractor maker with roughly 40% share, a dealer network no input company can match and a captive financier at the point of sale. It sits in this chain as the farmer's counterparty, and it captures more of the farm-gate wallet than any input supplier does. Against it: 15.1% ROCE is conglomerate-diluted and D/E 1.5 reflects the finance arm.
Excluded: Krishana Phoschem tops the screen on growth (66.7%) and return (27.2%), but SSP and phosphatics are subsidy-set commodity products with no pricing power — the return is a window, not a moat.
This is the weakest chain on the site, and the numbers say why: only 12 listed names, and the two best returns belong to companies that buy milk rather than process it. Procurement is a genuine moat and a poor business; brand is a good business and a fragile moat.
84.1% ROCE, 23.5% margin — among India's largest single procurers of milk, and it earns roughly five times what any listed dairy processor does on the same raw material. The moat is infant-nutrition regulation and Maggi, not dairy. Listing it here is the finding: in dairy the value is captured one stage past the milk. Against it: 11.1% sustained growth at 76x.
16.7% ROCE, 16.4% ROE, zero debt, 13.6% sustained growth at 27x — a tighter procurement geography than Hatsun (AP, Telangana, Karnataka) converted into materially better working capital. In a business where the whole game is milk cost per litre and days of inventory, geography discipline is the edge. Against it: small, and an African operation that adds risk without adding a moat.
22.0% ROCE, 18.7% margin, 24x — the best return of any listed dairy-format business, because ice cream carries brand pricing and a cold chain that acts as a barrier. Frozen distribution is expensive to build and expensive to copy. Against it: ₹5,316 Cr, family-governance history, and HUL's Kwality Wall's is now a separately listed competitor with deeper pockets.
The largest listed private milk procurer, buying direct from farmers through its own village-level network — the only genuine structural advantage in this chain, because it removes the cooperative intermediary. Against it: 15.2% ROCE and 11.2% growth at 59x. The moat is real; the price assumes it converts into returns it has not yet earned.
Excluded: Kwality Wall's (India) has no positive return history as a standalone yet — negative ROE, negative margin in its first reported period.
Defence is the chain where regulatory barriers are the moat, explicitly. The question is which barrier: a licence (explosives), a qualification (rotating parts), a nomination (warship yards) or a monopoly customer relationship (electronics). All four are durable; only some come with a margin.
36.4% ROCE, 28.6% operating margin, zero debt, 14.4% sustained growth at ₹3.0 lakh Cr. Radars, EW, communications and missile electronics with a near-monopoly on Indian defence electronics — programme cycles run 5–10 years, and the indigenisation mandate keeps imports out. It behaves like a high-quality industrial compounder that happens to be state-owned. Against it: one customer, and 49x for a business whose growth is a budget line.
38.1% ROCE, 27.3% operating margin, 12.4% sustained growth, profit compounding 30.4%. Explosives manufacture is licence-gated in a way that few industries are; Solar built the licence base in commercial mining explosives and has walked it into propellants, warheads and the Nagastra loitering munition. The defence business is a free option on a profitable industrial base. Against it: 92x.
43.0% ROCE — the highest on this map — with 39.8% sustained sales growth and no debt, at 37x. Four or five yards in India can build a warship, orders are nominated rather than tendered, and the order book is full. Against it: an 11.1% operating margin. This return comes from capital turns and customer advances, not from pricing power — a delivery slip hits it hard.
37.7% operating margin — the highest in components — with 33.8% sustained growth and profit compounding 63.1%. Rotating airfoils and turbine blades are qualified over multi-year cycles by energy and aerospace OEMs; once you are on the drawing, you are very hard to remove. Against it: ROCE is still only 11.9% while capacity is built, and 132x assumes the qualification pipeline converts on schedule.
Excluded: Sigma Advanced Systems tops the raw screen (60.8% ROCE, +1,037% TTM sales) entirely on a tiny base — the site's own note flags it. Zen Technologies is a real design-led franchise but is currently shrinking (sales −10%, profit −23%) at 94x.
One number explains this whole chain: Dixon earns 42% on capital at a 3.6% operating margin, and BEL earns 36% at 28.6%. Contract assembly makes money on capital turns; owning the design and the customer makes money on price. Both work. They are not the same business, and they should not be valued the same way.
29.2% ROCE on a 3.6% operating margin, with 49.9% sustained sales growth. The model in one line: the capital employed is small, the volume enormous, the margin thin by design. The moat is scale plus incumbency inside customer supply chains (JVs with global brands lock the relationship in a way a pure contract does not) plus PLI participation. Against it: 3.6% leaves no room for error, PLI has a sunset, and a handful of customers drive the volume.
36.4% ROCE at a 28.6% operating margin — the best margin-and-return combination in Indian electronics, because it owns the product design and the customer relationship instead of renting capacity to someone else's brand. This is what the rest of the chain is trying to become. Against it: a defence budget is its addressable market.
69.9% sustained growth, 37.5% ROCE — the clearest listed case of Indian design attached to Indian manufacturing, in AI servers and HPC. It sits one layer above pure EMS and earns a 13.3% margin for it, roughly four times Dixon's. Against it: 113x, and the customer base is narrow enough that one deferred order moves the year.
47.7% sustained growth, profit compounding 56.7% in high-mix industrial, automotive and medical EMS — stickier than consumer assembly because qualification per programme is real. The OSAT plant at Sanand is the option on Indian chip packaging. Against it: 12.7% ROCE, and the OSAT is being funded by equity raises rather than by the EMS business's own cash flow. Judge it on the core, treat the OSAT as free.
Also close: Syrma SGS — 33.0% sustained growth, rising ODM mix, but 16.8% ROCE at 78x.
India's highest returns on capital are on this map — 157% at P&G Hygiene, 108% at Colgate, 91% at Gillette — and they come with almost no growth. The interesting question in FMCG is no longer who earns the most; it is who is taking the margin, and the answer is increasingly the retailer.
38.9% ROCE, 33.6% operating margin, zero debt, at 17x with a 5% yield — the widest gap on this map between the quality of a business and the multiple applied to it. Cigarettes are a taxed duopoly, which is the most defensible position in Indian consumer goods precisely because the tax regime deters entry, and they fund a foods and paper business with real scale. Against it: 3.6% sustained growth and a persistent tobacco discount that has not narrowed in a decade.
34.5% sustained sales growth, 28.3% ROCE, profit compounding 68.7% — the only large name on this map with both growth and returns. Zudio's private-label vertical model means Trent owns the design, the sourcing and the shelf, and pays no brand royalty to anyone. That is a structural transfer of margin away from the brand owners above it. Against it: 86x, and same-store growth has been decelerating — this is priced for the model to keep working perfectly.
47.0% ROCE, 42.8% ROE, 11.1% sustained growth, TTM sales +25.1%, D/E 0.1. Category ownership in coconut oil (Parachute) and edible oils (Saffola) gives real pricing power, and the foods and digital-first portfolio is growing fast enough to matter. The best combination of quality and growth among the pure brand owners. Against it: 58x, and copra price swings drive the reported line.
23.4% operating margin, 19.7% ROCE, 18.1% sustained growth. Exclusive PepsiCo bottling territories are a contractual monopoly over geography, reinforced by cooler and route density that a challenger cannot replicate quickly. Against it: it manufactures someone else's brand, which is exactly why it earns a fifth of what Colgate earns on capital. The moat protects the territory, not the price.
Excluded: Colgate, Gillette and P&G Hygiene have the best returns in India and 3–11% growth; P&G Hygiene's profit fell 34% in the latest quarter. Eternal's 93.7% growth comes with 2.5% ROCE at 711x.
The pattern is stark and consistent: the businesses that charge for access — exchanges, asset managers, depositories — earn 40–115% on capital because they employ almost none. The businesses that lend earn 15–30% on equity and carry the credit cycle. Both can be good investments; only one has a moat that survives a downturn.
60.0% ROCE, 64.3% operating margin, zero debt, 52.0% sustained sales growth with profit compounding 69.4%. An exchange is a natural near-monopoly: liquidity attracts liquidity, and the licence keeps entrants out. Against it: 47x on volumes that SEBI can and does throttle — the derivative-position curbs of the last two years are the precedent, not the exception.
71.3% ROCE, 72.0% operating margin, 42.6% sustained growth — an effectively unchallenged position in Indian commodity derivatives. It owns no inventory and takes no position; it charges for access to liquidity that exists because it is there. Against it: concentration in a single asset class, a difficult platform-migration history, and 49x.
85.8% ROE, 74.3% operating margin, 21.9% sustained growth — a fee on other people's money with almost no capital behind it, distributed through ICICI Bank. Distribution is the moat in Indian asset management, and few distributors are larger. Against it: TER regulation is a permanent overhang on fee businesses, and the listing is recent (2025), so the public record is thin.
30.9% ROE, 22.0% sustained growth, at 10x earnings. Gold loans are short-tenor, over-collateralised and structurally resistant to credit cycles — this is the lending book that survives the downturns that kill unsecured lenders. Branch density in gold-loan catchments is a real barrier. Against it: 4.0x leverage, and RBI LTV and auction rules can reset the economics with a circular.
Excluded: Jio Financial screens first on growth (328%) — that is a spin-off base, not a growth rate, against 1.9% ROCE and 79x. Groww is a genuine franchise (37.3% ROCE, 59.6% growth) but its revenue is the same F&O volume SEBI is actively curbing.
Hospitals earn their return from occupancy on land they already own, which is why a single mature campus can out-earn a five-hospital network still ramping. Diagnostics splits in two: pathology is being discounted to the floor by aggregators; radiology cannot be, because it needs capital and radiologists.
26.6% operating margin, 22.4% sustained growth, D/E 0.3 — the highest revenue per occupied bed in the listed set, from metro capacity in NCR and Mumbai where beds per thousand are structurally short and new supply is planning-constrained. Growth is brownfield on land already owned, which is why margin holds as it expands. Against it: 14.7% ROCE while the expansion is in progress, at 64x.
42.2% operating margin — the highest in Indian diagnostics — with 20.5% ROCE and 16.7% sustained growth. The weighting to radiology rather than pathology is the whole argument: an MRI cannot be undercut by a discount aggregator with a phlebotomist and an app, because it needs the machine and the reporting radiologist. Against it: a south-India concentration and 82x.
22.6% ROCE, 27.9% margin, 15.8% sustained growth, at 26x — a single mature Coimbatore campus, and the clearest illustration that one fully-utilised hospital beats five ramping ones. Against it: single-asset, single-city. There is no growth story and the multiple says so.
28.0% ROCE, 29.0% operating margin, D/E 0.1 — the strongest referral network and brand in north-India pathology, which is the only durable defence against price-led entrants. Against it: 11.1% sustained growth at 56x. Quality is not in dispute; the price is.
Watch: NephroPlus is the most annuity-like clinical model listed here — dialysis is thrice-weekly and lifelong — with 28.4% sustained growth, but only 8.8% ROE at 82x. Manipal Health has the largest bed count and 29.1% growth at 12.1% ROCE and 106x.
Indian IT's problem is not returns — TCS earns 63% on capital — it is growth, and the market has repriced the whole sector as a yield asset. The names worth owning are where a product or a specialisation creates switching cost that a staffing contract does not.
45.3% ROCE, 50.4% operating margin, zero debt, at 30x with a 3.4% yield. Flexcube core banking has the highest switching cost in enterprise software — replacing a core banking system is a multi-year, bet-the-bank project that banks defer for decades. That installed base is the moat, and the 50% margin is what it is worth. Against it: 9.0% sustained growth, and Oracle's majority control means minority shareholders are along for the ride.
31.4% ROE, 31.5% ROCE, 33.9% operating margin, 42.0% sustained growth — the best economics in the services layer, from the narrowest specialisation: revenue-cycle management embedded in US healthcare provider workflow. Once you are running a hospital system's billing, you are not switched out casually. Against it: US healthcare policy risk, client concentration, and 41x.
34.4% ROCE, 20.9% sustained growth, profit compounding 27.6% — consistently the fastest-growing large mid-cap in Indian IT, with a genuine engineering-services position rather than a rebadged staffing book. Against it: 44x, and mid-cap IT growth premiums have historically compressed fast when the cycle turns.
63.0% ROCE, 26.9% operating margin, at 15x with a 2.6% yield. Extraordinary returns on almost no capital, the deepest client tenure in the industry, and the balance sheet to buy through a downturn. Against it: 5.8% sustained growth is the entire bear case, and it is a real one — the multiple is not a mistake, it is a judgement about AI-era pricing that may or may not be right.
Also close: Coforge grew 27.0% sustained at 23.5% ROCE, but the growth is acquisition-led (Encora) at 41x.
Building infrastructure is a bad business — contractors take execution and receivable risk for single-digit margins. Supplying it and owning it are good businesses. Every name below is a supplier or an owner; not one is a pure contractor.
33.2% ROCE, zero debt, 26.9% sustained growth, TTM sales +32.1% — roughly 30% of India's organised wires and cables, sold through a distributor and electrician network that is the actual moat: the specifier is a tradesman who trusts a brand, not a procurement department running a tender. It supplies metros, data centres, housing and railways, so no single capex cycle owns it. Against it: 49x, and copper is a pass-through that flatters and punishes the reported line.
31.8% ROCE at an 8.0% operating margin, D/E 0.1 — the fabricator that out-earns the mills supplying it, on direct-forming technology and the largest structural-tube distribution network in India. Low margin, high turns, and a product the mills cannot easily make themselves. Against it: 12.6% sustained growth and complete exposure to the steel price cycle at 47x.
57.9% operating margin, 22.9% sustained growth — concessions, land banks and roughly a quarter of India's port cargo on its own reporting. A port concession is the closest thing to a permanent monopoly in Indian infrastructure. Against it: 14.1% ROCE on very heavy assets, D/E 0.7, and a governance discount the market has applied consistently since 2023.
31.6% ROCE, zero debt, at 31x — the dominant share of India's mobile crane market, and the listed proxy for construction that is actually happening rather than being announced. Against it: 14.9% sustained growth and a pure capex-cycle beta. It will de-rate hard when ordering slows.
Note on the financing layer: PFC and REC earn ~20% on equity at 5–6x earnings and 6–7x leverage. That is a spread business on state credit, not a moat — the multiple is the market's view of the credit, not a mispricing.
The margin pool in Indian logistics sits almost entirely in the concession — ports earn 48–63% operating margins because the government granted them the right to be there. Everything downstream of the gate is a competitive service business earning single-digit to mid-teen returns.
57.9% operating margin, 22.9% sustained growth, the largest private port operator and the most integrated position in the chain — it owns the concession, the land behind it and increasingly the logistics that moves the box inland. Integration is what turns a port into a network. Against it: 14.1% ROCE, D/E 0.7, governance discount.
48.3% operating margin, 18.8% sustained growth, profit compounding 28.6% — the second listed private port group, anchored on captive JSW Steel and JSW Energy cargo while it builds third-party volume. Captive parentage guarantees the base load. Against it: that same captive parentage is the concentration risk, and 13.6% ROCE at 50x is a lot to pay for it.
62.5% operating margin, 28.1% ROCE, zero debt, at 15x — the purest read on port economics in India: one asset, no diversification to hide behind, APM Terminals' network behind it. Against it: 8.1% sustained growth, and the concession is finite. Concession renewal terms are the entire investment case; do not buy this without reading them.
19.4% ROCE, 19.2% ROE, D/E 0.1, at 15x — the best return of any non-port logistics name here, from multimodal operations where coastal shipping and rail undercut trucking on the same lane. Owning the assets on the modes that are structurally cheaper is a defensible position. Against it: 9.1% sustained growth — this is a well-run business in a low-growth format.
Excluded: Aegis Vopak tops the raw screen on profit growth (+1,472%) from a near-zero base, at 7.6% ROCE and 113x. Great Eastern Shipping at 5x is a shipping-cycle call, not a moat.
The honest finding here is that India has no high-return listed pure-play recycler at scale. The best economics on this map belong to primary miners and to the offtakers who buy secondary metal — which is exactly what you would expect from a chain where roughly nine-tenths of scrap still moves through informal aggregators.
24.1% ROCE, 23.9% sustained sales growth, profit compounding 41.2%, D/E 0.2, at 25x. Hazardous-waste and lead-recycling licences are the barrier — you cannot open a secondary lead smelter because you want to — and Pondy is extending the same licensed collection network into copper, aluminium and zinc. Against it: a 6.9% operating margin. Recycling is a spread business on scrap-to-metal, and the spread is not proprietary.
30.4% ROE, 25.7% ROCE, 46.0% sustained growth — among the largest listed non-ferrous recyclers by revenue, across lead, copper and aluminium. Scale in scrap sourcing is the only real advantage available in this business. Against it: 5.4% operating margin, D/E 0.8, and it listed in October 2025 — there is barely a public record.
17.0% ROCE, 15.1% sustained growth, D/E 0.3 — the broadest collection and aggregation network among listed recyclers, across geographies and metals. In a chain whose central problem is fragmented supply, the collection network is the asset. Against it: 9.4% margin and 34x — the market is paying a growth multiple for a spread business.
The finding: the two best sets of economics on this map are HBL Engineering (58.5% ROCE, 32.0% margin, zero debt, 23x) and Hindustan Zinc (69.2% ROCE, 56.0% margin) — an offtaker and a primary miner. Neither recycles anything at scale. Until formalisation forces volume through licensed hands, that is where the value stays.
One rule governs this entire chain: the money is in the orebody, not the mill. Every name below owns a resource or is integrated back to one; the non-integrated processors earn a fraction of these returns on far more capital.
69.2% ROCE, 56.0% operating margin, at 14x with a 1.8% yield. Rampura Agucha and the Rajasthan zinc-lead complex sit in the first quartile of the global cost curve and produce roughly three-quarters of India's zinc. A first-quartile orebody is the most durable moat in commodities — it is the only one that works when the price falls. Against it: 6.0% sustained growth, zinc-price beta, and a parent whose dividend needs have driven balance-sheet decisions before.
39.6% ROCE, 47.3% operating margin, zero debt, at 11x with a 3% yield — captive bauxite at Panchpatmalli feeding its own alumina and smelter. That integration is precisely why its returns beat non-integrated smelters buying alumina at market. Against it: 7.8% sustained growth and full exposure to the alumina cycle.
27.6% ROCE, 28.8% margin, 15.8% sustained growth, at 10x with a 3.9% yield — India's largest iron ore miner and the effective domestic price setter. Owning the orebody is the whole business. Against it: state ownership cuts both ways — pricing has been leaned on for policy reasons before, and can be again.
72.2% sustained sales growth — the fastest in the chain — at 27.3% ROCE and 36.8% margin, 23x. The Surjagarh orebody coming online is a genuine step-change in supply, not an accounting effect. Against it: D/E 1.5 and a large capex programme into pellets and steel. The growth is a ramp; the question is what the return looks like once the capital is fully deployed downstream.
Excluded: Coal India earns 35.0% on capital at 8x but is shrinking (−0.7% profit CAGR) and produces thermal, not the coking coal Indian steel actually needs.
Two distinct good businesses live in this chain: branded domestic formulations, where prescriber loyalty means near-zero capital and no USFDA risk, and CRDMO, where a top-25 pharma relationship takes five to eight years to build and is not tendered annually. Plain generics are neither.
44.8% ROCE, 28.1% operating margin, zero debt, at 35x. Branded domestic formulations in gastro and women's health — the prescriber writes the brand, the capital intensity is near zero and there is no export or regulatory-inspection risk at all. This is the cleanest business model in Indian pharma. Against it: 9.0% sustained growth. You are buying certainty, not compounding.
29.9% ROCE, 28.7% ROE, 31.8% operating margin, profit compounding 37.0%, at 17x. Complex inhalation generics are the hardest category to copy — device plus molecule plus bioequivalence — which is why they behave like limited-competition products rather than commodities. The multiple has not caught up with the operating turnaround. Against it: US generic price erosion is relentless, and a small number of products carry a large share of the profit.
29.2% operating margin, 21.7% sustained growth, profit compounding 42.2% — discovery-to-commercial CRDMO working with a large share of the global top 25. The moat is relationship tenure and regulatory filings that name the site: switching a validated manufacturer mid-programme is not done casually. BIOSECURE-driven China substitution is a genuine, if slow, tailwind. Against it: 84x.
39.6% operating margin — the highest in the chain — with 30.4% ROCE and zero debt. Fermentation-based capability plus specialty ingredients is a differentiated technology base, not just capacity for hire. Against it: listed in 2025 with a soft first public quarter (TTM sales −2.1%) at 84x. The record is one year long.
Also close: GSK Pharma earns 61.4% on capital with a 2.2% yield, but at 5.5% sustained growth. Neuland's spectacular quarter (+962% profit) is milestone-driven CMS revenue and does not annualise.
This chain has two very different opportunities. Grid hardware is a genuine bottleneck with 45–77% returns on capital and multi-year lead times. Solar manufacturing has the growth — 58–62% sustained — but its barrier is ALMM, which is policy, not moat. Price them differently.
77.4% ROCE, 26.2% operating margin, zero debt — the highest return on this map, from high-voltage substations and grid integration against a national transmission rebuild. Near-zero capital employed, multi-year certification, three credible suppliers. Against it: 82x, and this is an order-book business.
33.3% ROCE, 30.0% operating margin, 62.0% sustained growth, profit compounding 132.7%. Integrated cell and module manufacture — cells are the harder, more capital-intensive half of the chain and where ALMM protection bites hardest. Being on the cell side is the difference between a manufacturer and an assembler. Against it: D/E 0.9, and the protection that creates the margin is a government list that can be revised.
38.8% ROCE, 57.8% sustained growth, TTM sales +94.3%, at 19x — India's largest module maker, integrating backward into cells and wafers, with US export exposure. It is the cheapest large name in the renewable complex because the market prices modules as a commodity. That is the right instinct; the integration is the counter-argument. Against it: module prices globally are still deflating, and the ALMM dependency is the same as Premier's.
50.7% ROCE, 25.5% operating margin, zero debt, 32.5% sustained growth, at 33x — an Indian transformer maker winning export order flow, particularly into renewable projects. That it wins on global tenders rather than only domestic ones is the evidence that matters. Against it: ₹4,487 Cr, TTM sales −7.0%, and a single plant.
Excluded: NTPC Green (156% growth, 3.6% ROCE, 128x) and Diamond Power (+401% growth off a post-insolvency base) are both base effects. Suzlon is a real turnaround — 35.1% ROCE, 38.0% growth, D/E 0.1 — but the dilution and restructuring history argues for a smaller position, not a top slot.
Restaurants are the worst business on this map. Packaged food brands earn 22–84% on capital; the eight listed restaurant operators earn a median 4.4%, from −0.5% to 14.8%. The two positions worth owning are the brand that sits on the shelf and the platform that owns the ordering occasion.
84.1% ROCE, 23.5% operating margin — the highest return on this map by a distance, from confectionery, dairy, beverages and prepared foods. The gap between this and the restaurant operators' median 4.4% is the argument of the entire chain. Against it: 11.1% sustained growth at 76x.
23.4% operating margin, 19.7% ROCE, 18.1% sustained growth — the beverage on the tray is the highest-margin line in any QSR order, and Varun bottles it under exclusive PepsiCo territories. At ₹1.47 lakh Cr it is worth close to twice the entire listed restaurant universe on this map. Against it: very heavy capital in plants, glass, crates and coolers, and it does not own the brand it sells.
22.0% ROCE, 13.2% sustained growth, profit compounding 25.2%, D/E 0.1 — the third-largest ethnic snacks player, competing for the same snacking occasion the QSR chains chase, from the shelf rather than the counter, at a fraction of their capital intensity. Against it: 55x for mid-teens growth, and namkeen is a regionally fragmented, price-competitive category.
93.7% sustained growth, and the only company here that owns the customer relationship across food delivery, quick commerce and B2B supply. Blinkit's dark-store density is the most defensible new moat built in Indian consumer in a decade — location coverage compounds and cannot be bought quickly. Against it: 2.5% ROCE and 711x earnings. The position is the best on the map; nothing in the current returns supports the price. Own it for the position, and size it for the fact that the economics are entirely prospective.
The finding: Eternal alone is worth roughly four times every listed restaurant company on this map combined. The chain's value has moved from the kitchen to the shelf and the app.
Railway capex is a government programme, so the durable positions are the ones gated by an approval: Kavach vendor certification, a statutory ticketing monopoly, or exclusive right-of-way. Contractors bidding for electrification work have none of these.
58.5% ROCE, 32.0% operating margin, zero debt, 29.4% sustained growth, profit compounding 104.7%, at 23x. One of very few approved Kavach suppliers — approval is the barrier, and the approved list is short. Best economics on this map, and the multiple has not run away from them. Against it: the revenue is a government programme's rollout schedule, and rollouts slip. Order lumpiness is structural.
46.1% ROCE, 30.5% operating margin, zero debt, at 29x — a statutory monopoly on rail e-ticketing, plus catering and tourism, running one of India's highest-transaction consumer platforms. Against it: the same statute that grants the monopoly sets the convenience fee, and it has been changed by announcement before. That single policy line is most of the risk.
22.9% ROCE, zero debt, 26.2% sustained growth, at 24x — exclusive right-of-way for optic fibre along the rail network, plus a growing government systems-integration book layered on top. Right-of-way along the tracks cannot be replicated at any price. Against it: PSU margin discipline is patchy and the systems business is a low-margin add-on to a good asset.
36.4% ROCE, 28.6% margin, zero debt — railway signalling, communications and security on top of the defence electronics base, so a slip in the rail programme does not break the year. Against it: railways are a small share of the revenue; you are buying the defence business with a rail option.
Excluded: Kernex Microsystems tops the screen (88.0% growth, +305% TTM) but is a micro-cap that moves on programme announcements rather than delivered revenue. Hitachi Energy India is a fine business at 133x. Data fix: the Transrail Lighting entry on this map was pointing at TARIL (Transformers & Rectifiers India) and has been corrected to TRANSRAILL.
Development is a working-capital business dressed up as an asset business; the returns come from land bought cheaply long ago, or from lending against the finished product. The most reliable earner in the chain is neither — it is the supplier selling into every project regardless of who builds it.
32.2% operating margin, 20.8% sustained growth, profit compounding 56.5%, D/E 0.4, at 30x — the largest listed developer by pre-sales, with an MMR land bank carried at historical cost and a shift toward joint-development agreements that grows the book without buying the land. Brand premium in residential is real and measurable in realisation per square foot. Against it: 16.4% ROCE, and residential development is the most cyclical business in this chain — pre-sales momentum is the whole story and it turns quickly.
20.1% ROE, 26.1% sustained growth, at 13x — affordable housing finance to self-employed borrowers in tier-2 and tier-3 south India, a segment banks will not underwrite because it has no salary slip. The underwriting capability and branch density are the moat; the yields follow from it. Against it: D/E 1.6 and a borrower base with no formal income record — this book has never been through a real stress cycle at this size.
33.2% ROCE, zero debt, 26.9% sustained growth — wires and cables into essentially every building constructed in India, roughly 30% of the organised market, sold through electricians who specify by brand. It gets paid whoever wins the plot. Against it: 49x.
7.8% yield, 71.3% operating margin — the only institutionally managed Indian REIT without a developer sponsor, which removes the conflict that sits inside every sponsor-managed REIT: the developer decides what to sell into the trust and at what price. Against it: 5.4% ROCE and 2.8% ROE — a REIT is a yield instrument, and office absorption is the variable that matters.
Also close: Home First Finance grew 31.5% sustained but at 2.4x leverage; Bajaj Housing has the distribution and 4.6x leverage at 26x.
No sugar mill appears in this list, and that is the finding. The 24 listed mills on this map earn a median 8.0% on capital converting cane into a price-controlled commodity; Britannia earns 56.0% converting that sugar into biscuits. The entire listed sugar-mill complex is worth ₹72,317 Cr — less than half of a single bottler on the same map.
24.1% ROCE, 18.3% operating margin, D/E 0.1, 20.5% sustained growth — a Rampur molasses distillery that became a premium IMFL house (Rampur single malt, Jaisalmer gin, Magic Moments). Two moats stack here: state excise licensing keeps entrants out of each market individually, and premium brands price on aspiration rather than input cost. It is backward-integrated into its own distillery, so it captures the spread the mills give away. Against it: 89x, and every state excise policy is an independent political risk.
26.4% ROCE, 18.2% operating margin, D/E 0.1 — the largest prestige-and-above spirits franchise in India, with Diageo's portfolio and a franchising strategy that has shed the low-margin popular tail. Against it: 5.5% sustained growth. The premiumisation trend is real but slow, and the multiple (60x) assumes it accelerates.
22.2% operating margin, 18.0% ROCE, 16.3% sustained growth, profit compounding 50.3% — a Haryana sugar mill that turned its own distillery into Indri, a genuinely differentiated single malt, and re-rated to more than triple the value of distillery-heavy peers. It is the clearest proof that the escape route from commodity sugar is a brand, not more crushing capacity. Against it: ₹6,969 Cr resting on one brand, D/E 0.6, and 50x.
23.4% margin, 19.7% ROCE, 18.1% sustained growth — among India's largest industrial buyers of refined sugar, and worth more than twice every listed sugar mill combined (₹1.46 lakh Cr against ₹72,317 Cr). It sits on this map as the customer, and it captures the margin the miller cannot. Against it: it is a bottler, not a brand owner, which caps the return at roughly a fifth of a Nestlé or a Britannia.
Watch: Allied Blenders approved a 66 million bulk-litre distillery at Moradabad in August 2026 — a major ENA buyer integrating backwards, which is a direct threat to the mills currently supplying it.
This chain offers three defensible positions and no more. A consolidated three-player market has restored pricing power to the operators; the passive infrastructure beneath them is an annuity with one large receivable problem; everything else on this map is an IT company or a contract manufacturer that happens to serve telecom.
56.9% operating margin, 20.3% ROE, 17.6% ROCE, 14.9% sustained growth, with borrowings down ₹18,000 Cr over a year. Three operators, rational tariff behaviour and ARPU leadership is as close to an oligopoly moat as Indian telecom has ever offered — and Airtel has the best subscriber mix, the enterprise business and Africa alongside. Against it: 39x, D/E 1.3, permanent spectrum capex and an AGR liability that is a policy decision away from moving either way.
25.9% ROE — higher than its parent's — with a 52.6% operating margin, 12.4% sustained growth and profit compounding 34.5%. Airtel's Rajasthan and North East circles listed separately: the purest available expression of tariff repair, without the Africa and enterprise businesses diluting the signal. Against it: it trades at a premium to the parent that owns it, which is difficult to justify on anything but scarcity.
54.6% operating margin, 19.5% ROCE, at 14x — tower portfolios are long-tenor contracted annuities with switching costs measured in years, and this is the largest listed one. Against it: 4.6% sustained growth, and Vodafone Idea receivable exposure is not a theoretical risk — it is the reason the multiple is 14x and not 25x. The moat is genuine; the customer is the problem.
Excluded: Tata Teleservices' apparent 55.6% ROCE is an artefact of negative net worth (book value −₹102/share), not a quality signal — the site flags this on the map itself. For enterprise ICT exposure, Infosys (40.0% ROCE, 4.1% yield, 15x) and TCS are the better vehicles.
Water is a government-budget business, and that shapes everything: the good positions are the ones with a technical specification behind them or an annuity contract attached, and the dangerous ones are those whose entire order book is a single subsidy scheme.
21.3% ROCE, near debt-free, India's largest listed pure-play water company across drinking water, wastewater, industrial water, desalination and reuse, executed through EPC, O&M, DBO, BOOT and HAM. The O&M and BOOT books are annuities, which is what separates it from a contractor. Process IP and a global desalination reference list are real barriers in a business won on technical qualification. Against it: 6.8% sustained growth and municipal receivables — the customer is a state utility, and state utilities pay late.
23.6% ROCE, 23.8% sustained growth, profit compounding 27.8%, D/E 0.3, at 29x — in-house motors and controllers make it the most integrated solar pump maker, and integration is why its margin beats assemblers'. Against it: be blunt about this — the demand is PM-KUSUM. That is a subsidy programme with a budget and an end date, not a moat. The manufacturing capability survives the scheme; the growth rate does not.
17.7% ROCE, 19.7% sustained growth, D/E 0.2 — water control gates, screens, penstocks and intake equipment: the physical interface between a river and a treatment plant. These get written into municipal specifications by name, which is a switching cost most equipment makers never achieve, and the export book proves the products compete internationally. Against it: ₹3,052 Cr and 35x for a business whose orders are municipal tenders.
20.7% ROCE, zero debt — India's largest plastics processor, with piping as the biggest division, riding both housing and government water schemes through a distribution network built over decades. Against it: 6.8% sustained growth and a slightly negative profit CAGR. This is a steady compounder having a poor patch, at 44x.
Watch: Oswal Pumps screens highest in the chain (75.0% sustained growth, 38.2% ROCE, ~10x) but listed in 2025 with the overwhelming majority of revenue from PM-KUSUM. The multiple tells you what the market thinks that concentration is worth.
Not investment advice. These are positional and quality judgements built on one snapshot of trailing accounting data. Trailing ratios say nothing about what is already priced in — several of the strongest businesses here trade on multiples that assume the moat holds and the growth continues. Verify every figure before acting on it.