How to Analyse a Specialty Chemicals Stock

In the December 2025 quarter, Deepak Nitrite's Advanced Intermediates business did everything a growth investor wants: revenue reached ₹652 crore, up 18% year-on-year and 11% quarter-on-quarter, "driven by higher volumes with improved market penetration of key products" (Deepak Nitrite Q3 FY26 concall, 13 February 2026). And the profit that volume threw off? Segment EBIT was ₹15 crore — "reflecting this continued pricing pressure arising from aggressive Chinese dumping and global oversupply" (same call). A fifth more volume, and the money it earned all but disappeared. (Illustration of how to read the numbers, not a view on the stock.)

That single contrast is the whole subsector in one line. A specialty chemicals company sells tonnes, but it earns on the gap between what it pays for raw material and what it gets for the finished molecule — the spread. Volume is the thing you can see in the press release; the spread is the thing that decides whether the volume was worth making. Read the two as one number and you will mistake a margin collapse for a growth story, which is exactly what the Q3 FY26 Advanced Intermediates line invited you to do.

This is how to read a specialty chemicals company the way a sector analyst does: the handful of operating numbers that actually decide the outcome — most of them buried in investor decks and concall transcripts, not the income statement — what "good" looks like for each, the right valuation lens, and the one risk that has quietly halved more chemical multiples than any demand shock.

A note on the boundary first. From the outside you will not model a plant's per-product economics. What you can do is separate volume from spread, watch where capital is going and what it earns, and check whether management's account of the cycle survives the next two quarters.

What actually drives the economics here?

A specialty chemicals firm is a conversion business. It buys a basic feedstock — a derivative of crude, fluorspar, benzene, ammonia — runs it through a proprietary process, and sells a higher-value molecule to a customer who needs that exact specification: an agrochemical innovator, a pharma company, a refrigerant blender. The moat, where one exists, is process know-how and a qualified relationship — the customer has spent years validating your plant, and switching is slow and risky for them.

That framing fixes the most common beginner error: treating chemicals like a fast-moving consumer business where revenue growth is the headline. It isn't. It is closer to a refinery with a chemistry degree — the asset is capital-heavy, the product is half-commodity, and the difference between a great year and a terrible one is often a few rupees of spread per kilo that you never see in the topline. The whole job is to find the businesses where the spread is structural (protected by process, scale, or a captive customer) and avoid the ones where it is cyclical (set by global supply, and therefore by China).

The metrics that matter — and where they hide

Most of these are not in the profit-and-loss statement. They live in the investor presentation and the concall Q&A. That is precisely why they are worth the effort: anyone can read the topline.

Volume growth, read separately from price

What it is: tonnes shipped, stripped of price and currency. Why it matters here above all: revenue can rise on price and fall on volume in the same quarter, or the reverse — and the two have opposite meanings. Volume up with spread holding is a healthy share-gain; revenue up only because a feedstock got expensive (and got passed through) is no real growth at all.

Where to find it: almost never in the P&L — management discloses it on the call. Aarti Industries said its portfolio "achieved approximately 17% YoY volume growth in FY25" with energy-application volumes up 21% quarter-on-quarter and the base business up 14% (Aarti Industries Q4 FY25 concall). What "good" looks like: double-digit volume growth with a stable-or-rising spread. Volume growth bought by cutting price is the trap — and it looks identical to the real thing on the revenue line.

RM pass-through and the spread

What it is: whether the company can move its selling price when its raw-material cost moves — and how much margin it keeps in between. Why it matters: in a contract or formula-priced book, RM is passed through and the spread is defended; in a spot-priced commodity book, the company eats the swing. The Deepak Nitrite example above is the spot end of the spectrum — Advanced Intermediates topline grew 18% while EBIT fell to ₹15 crore, because the spread, not the volume, was set by Chinese oversupply (Deepak Nitrite Q3 FY26 concall). The same company's other engine, Phenolics, walked its EBIT margin from 8% in Q1 FY26 to 11% in Q2 FY26 (Inve data, Q1–Q2 FY26) on "better net raw material realization" — pass-through working in reverse, as a tailwind. Where to find it: the margin trend across six to eight quarters, cross-read against the management's words on pricing. A gross or EBIT margin sliding while volume rises is the spread leaking; that is the number to distrust.

Capex and asset turns — does the new plant earn its keep?

This is the metric that separates value-creators from value-destroyers, and it is the one most often buried. What it is: revenue (or, better, the segment's revenue) divided by the gross block it sits on — how many rupees of sales each rupee of plant generates. Why it matters: specialty chemicals grows by building plants, and a plant is dead money until it fills. A company guiding ₹1,000 crore of annual capex needs you to ask what asset turn that capital will earn, because a 1x-turn plant and a 2x-turn plant are different businesses wearing the same balance sheet.

Where to find it: management sometimes discloses it by segment. Navin Fluorine put numbers to it: roughly 2x asset turn for its CDMO and High-Performance Products, against 1.45x for base Specialty Chemicals (Navin Fluorine Q4 FY25 concall). The gap is the whole capital-allocation case — the same rupee of plant earns far more in CDMO. What "good" looks like: a rising asset turn as a new plant ramps from commissioning to full utilisation, and capex steered toward the higher-turn segments. The red version: capex going up, asset turns going nowhere, and free cash flow permanently negative. Aarti's FY25 capex was ₹1,372 crore and the FY26 guide moved up from "around ₹1,000 crore" to "about ₹1,100 crore" (Aarti Industries Q4 FY25 and Q3 FY26 concalls) — fine if the turns follow, a slow bleed if they don't.

Capacity utilisation — the bridge between capex and earnings

What it is: how full the plants are. Why it matters: a half-empty plant carries full depreciation and fixed cost, so utilisation is the dial that turns capex into margin. Where to find it: the call. Navin Fluorine told analysts it expected utilisation to rise from "50%, 60% this year" to "about 70%, 75% in the coming year," with FY27 visibility "almost up to about 80%" (Navin Fluorine Q4 FY26 concall). Read against its margin (more below), that utilisation ramp is the earnings story. What "good" looks like: a clear, dated path from low utilisation to high, with the margin moving in step.

Export mix — who is the customer, and where?

What it is: the share of revenue earned abroad, usually for innovator agrochem and pharma intermediates. Why it matters: export-heavy books are tied to global crop cycles, innovator R&D pipelines, and currency — higher quality of customer, but also exposed to global destocking and tariff shifts. Where to find it: the deck or the call. Aarti Industries reported export revenue share of 65% in Q3 FY26 and 55% in Q4 FY25, with the US alone at 15–20% of revenue (Aarti Industries Q3 FY26 and Q1 FY26 concalls). What "good" looks like depends on the model — but a high export share makes the tariff and destocking questions central, not optional.

The contract / CDMO pipeline — the closest thing to a real order book

What it is: in pharma-facing custom synthesis (CDMO/CRAMS), the book of molecules a company is making for an innovator, split between early-stage (small, lumpy) and late-stage/commercial (large, durable). Why it matters: a late-stage commercial molecule is the chemical equivalent of an annuity — qualified, sticky, high-turn. Where to find it: only the call, and management is often guarded. Navin Fluorine said it was "working close to about 50, 55 molecules, half of them being in late-stage commercial and half of them being in early stages" (Navin Fluorine Q4 FY26 concall). What "good" looks like: a pipeline tilting toward late-stage, because that is what de-lumps the revenue. The honest reader notes the limit too — on the same calls, management explicitly declined to disclose asset turns or contract values on confidential molecules, so part of this you take on management's word.

Gross margin — the spread, made visible

What it is: revenue minus raw-material cost, as a percentage. Why it matters here specifically: because RM is the biggest line, gross margin is the cleanest proxy for the spread you cannot see directly. A widening gross margin while feedstock prices are flat-to-down is real spread expansion; a gross margin holding only because prices were passed through tells you less. Where to find it: derivable from the P&L, but the why is on the call. Navin Fluorine's operating EBITDA margin tells the story plainly — 18.3% in Q4 FY24, 25.5% in Q4 FY25, with full-year FY25 at 22.7% versus 19.3% the year before (Navin Fluorine Q4 FY25 concall) — and the income statement backs it: quarterly OPM ran 19% in Q1 FY25 and 34% by Q4 FY26 (Inve data, Q1 FY25–Q4 FY26).

How should you value a specialty chemicals stock?

The instinct is to reach for P/E. Resist it at the cycle's extremes. Specialty chemicals earnings swing with the spread, so a P/E computed on a trough year looks punishingly high and a P/E on a peak year looks deceptively cheap — the multiple does the opposite of what you want at the moment you most need it. The more useful pair is P/E and EV/EBITDA read together, on normalised earnings, against where the spread sits in its cycle. EV/EBITDA has the virtue of being capital-structure neutral and less distorted by the depreciation surge a fresh plant brings — useful precisely when a company is mid-capex.

The number that should govern the multiple you pay is the asset turn the next wave of capex will earn. A company building 2x-turn CDMO plants deserves a richer multiple than one building 1.45x-turn commodity capacity, because the same rupee of capital compounds faster — Navin Fluorine's own segment turns make that case explicitly (Navin Fluorine Q4 FY25 concall). And the discount that should be applied is the China question. If a product's spread is set by Chinese supply rather than by the company's own process — the Deepak Nitrite Advanced Intermediates situation, where ₹652 crore of growing revenue threw off ₹15 crore of EBIT under "aggressive Chinese dumping and global oversupply" (Deepak Nitrite Q3 FY26 concall) — then no multiple on peak earnings is safe, because the earnings themselves are on loan from a supply cycle. Pay up for structural spread and high asset turns; pay down for spread that China can reset.

A worked case: when guidance walked the *right* way

Most of our concall work catalogues guidance that quietly went the wrong way. Navin Fluorine through FY25–FY26 is the rarer, more instructive case: guidance that kept moving — in the holder's favour — because the operating numbers underneath it kept improving.

Start with what management said, in sequence. In Q1 FY26 the chairman's team set a deliberately cautious bar: "I think there is reasonable confidence to say we'll be north of 25%" on EBITDA margin (Navin Fluorine Q1 FY26 concall). By Q2 FY26 that had been revised up to a 28–30% band; by Q3 FY26 the long-term annualised margin guidance was raised to 30% (Inve data, Q2–Q3 FY26). Now check it against what the company did. Reported OPM ran 29% in Q1 FY26 and reached 34% by Q4 FY26 (Inve data, Q1–Q4 FY26); net profit went from ₹51 crore in Q1 FY25 to ₹213 crore in Q4 FY26 (Inve data, Q1 FY25–Q4 FY26) as utilisation climbed toward the 70–75% the company had guided (Navin Fluorine Q4 FY26 concall).

Here is the discipline the case teaches. A "north of 25%" guide that becomes a "raise to 30%" guide is only worth trusting because the utilisation ramp and the segment asset turns made it mechanical — a low-utilisation fluorine plant filling up should throw off rising margin, and it did. The said-vs-did test is not about catching liars; it is about checking whether the operating bridge between the words and the numbers actually exists. Here it did. The investor's edge was not believing the optimism — it was verifying the utilisation and turn data that made the optimism arithmetic rather than hope. (Illustration, not a view on the stock; a read on how management communicated through one up-cycle, not a lifetime verdict.)

Red flags specific to this subsector

• Revenue growing while the segment's profit shrinks. The Deepak Nitrite Advanced Intermediates pattern — ₹652 crore of revenue up 18% on higher volumes, ₹15 crore of EBIT (Deepak Nitrite Q3 FY26 concall). Growing tonnes into a collapsing spread is not growth; it is paying to keep a plant busy.
• Capex rising while asset turns and free cash flow do not. A plant that never fills, or fills at a 1x turn, is a permanent drag dressed as expansion. Always ask what turn the new capital will earn.
• A spread set by China, sold as a moat. If the product is a global commodity intermediate, the spread is a supply-cycle rental, not a franchise. "Aggressive Chinese dumping and global oversupply" is the phrase to listen for (Deepak Nitrite Q3 FY26 concall) — and it can hit any Indian producer of a non-differentiated molecule.
• Margin held up only by inventory gains or pass-through, not by spread. A good gross-margin quarter that came from a feedstock move, not from the company's own pricing power, will reverse.
• A CDMO pipeline that stays "early-stage" forever. Lumpy early-stage molecules that never graduate to late-stage commercial mean the annuity never arrives. Watch the mix shift, not just the molecule count.

Where this lens can be wrong

The strongest case against everything above is that it rewards hindsight and punishes the cycle's bottom. Read strictly, "avoid spread set by China" would have kept you out of every commodity-exposed chemical name at exactly the trough — and troughs are where the multi-year returns in this sector are actually made, because spreads mean-revert and the survivors come out with more share. A reader who always discounts a China-exposed book will miss the best entry the sector offers; the Deepak Nitrite Advanced Intermediates line that looked like a red flag in Q3 FY26 is, on a different reading, a cyclical low that a patient owner buys because the spread is depressed. The honest claim is narrower than it sounds: separating volume from spread tells you what kind of earnings you are buying — structural or rented — not when the rented kind turns. It lowers your odds of mistaking a cyclical peak for a franchise. It cannot time the cycle, and it cannot tell you whether a given plant's process is genuinely hard to copy — that takes domain knowledge we are honest about not having from a transcript alone.

Frequently asked questions

A specialty chemicals company is a balance sheet that turns a feedstock into a spread, and almost every way it can disappoint shows up first in the spread — not the topline that the spread is hiding behind. Inve's KPI Screener lines up volume growth, margin trend, and capex across chemical names with a data-confidence flag per number, and the concall summaries pull every forward commitment into one guidance table per quarter, so the work in this guide takes minutes rather than an afternoon of transcript-reading. For the financing-side cousin of this discipline, see how to analyse an NBFC — same instinct, different balance sheet.

So invert the question you bring to a chemical company's results. Don't ask "is revenue growing?" Ask: if this company's spread were quietly being set by global oversupply rather than its own process, what would the numbers look like — and does this book rule that out? Growing volume into a shrinking segment EBIT does not rule it out; it is the pattern itself.

And the owner's question, the one to sit with before buying a share: five years out, when this cycle's spreads have come and gone, is this a business that makes its margin from a process competitors can't easily copy — or one that merely rents its margin from a supply cycle it doesn't control? If the honest answer leans on the cycle rather than the chemistry, you've bought the spread, not the company.

Related reading

• How to Analyse a Bank Stock (NIM, CASA, GNPA, ROA)
• How to Analyse a Building Products Company (Pipes, Tiles)
• How to Analyse a Cables and Wires Stock