How to Analyse a Diagnostics Stock (Volume, Realisation)

Here are two numbers from the same industry, from recent quarters. Dr Lal PathLabs earned ₹956 of revenue from the average patient who walked through its door (Revenue per patient, Q4 FY26 — Inve data; the company's own quarterly disclosure). Thyrocare earned ₹36 per test (revenue per test ex-Polo and Vimta, ₹36 in Q3 FY26 — Thyrocare Q3 FY26 concall). Same business — drawing blood, running it on an analyser, sending back a report. A 26-times gap in what looks like the "price" of the product. (Illustration of how to read the numbers, not a view on either stock.)

That gap is the whole sector in one comparison, and it is also the trap. The two numbers are not measuring the same thing. A patient is not a test — a patient buys a panel of tests, often a bundled wellness package, and pays retail. A test sold wholesale to a franchisee lab that then marks it up to its own patient is priced at a fraction of that. Read the two numbers as if they were comparable and you will conclude one company has collapsing pricing and the other has a moat. Read them correctly and you learn the most important thing about a diagnostics business before you look at a single margin: who is the customer, and at what point in the chain does this company capture the rupee?

This is how to analyse a diagnostics company the way an operator does — the handful of numbers that decide the outcome, where they hide (some are on the income statement; the ones that matter most are buried in the investor deck and the concall), and the one red flag that separates a real network from a subsidised land-grab. A boundary first, the way we'd flag it for any sector: you will not model a lab chain's catchment economics from the outside. What you can do is read the direction of volume, realisation and mix, and check whether management's account of them survives the Q&A.

What actually makes a diagnostics company money?

Strip a lab chain down and it is a volume business with a fixed-cost spine. The reference lab, the analysers, the logistics fleet that moves samples overnight, the senior pathologists — those costs barely move whether you run 40 million tests a quarter or 50 million. So the entire economic engine is operating leverage: push more samples through the same fixed plant, and a large share of each incremental rupee of revenue drops to EBITDA.

That single fact governs everything else. It explains why diagnostics companies obsess over collection-centre density (more touchpoints feed more samples into the same lab). It explains why a price war is so dangerous (cut realisation and the fixed cost doesn't shrink with it). And it explains why the headline you should distrust most is revenue growth on its own — a chain can grow revenue 15% by acquiring a loss-making regional lab, or by discounting to win B2B contracts, neither of which tells you whether the underlying machine is getting more efficient. The question is never "did revenue grow?" It is: did volume grow, did realisation hold, and did the mix improve — and did all three together show up as margin?

The metrics that matter (and where they hide)

Test and sample volume — the engine, not the headline

Volume is the number of tests or samples processed. It is the truest measure of how much work the fixed plant is actually doing, and it is almost never on the income statement — you dig it out of the investor presentation or the concall. Thyrocare processed 49.6 million tests in Q3 FY26, up 22% year-on-year (Thyrocare Q3 FY26 concall). Dr Lal's sample-volume growth was 12.9% in Q4 FY26 (Inve data, Q4 FY26).

Why it matters here specifically: because of operating leverage, volume growth that outruns revenue growth tells you realisation is softening (more tests, less money each), while revenue growing faster than volume tells you mix or price is doing the lifting. "Good" is mid-to-high single-digit organic volume growth compounding quarter after quarter — the steady drip that fills the fixed plant. A volume spike you should interrogate, not celebrate: Dr Lal itself flagged that a chunk of one period's growth came from a heavy seasonal-fever season, and when fever normalised the next year, "patient volume growth slowed to 2.7% due to an unexpected decline in the seasonal fever portfolio" (Inve data, Q3 FY26). Volume borrowed from a flu season is not volume you can compound.

Realisation per test (or per patient) — read which one, and why

Realisation is revenue divided by the unit — but which unit decides what you're reading. Revenue per patient captures how much the chain extracts from each person, which rises when they buy bundled panels and premium tests. Revenue per test captures raw price, which barely moves. Dr Lal's revenue per patient climbed from ₹861 in Q3 FY25 to ₹956 in Q4 FY26 (Inve data) — and management was explicit that this was not a price increase: "We have not had any price increase impact, rather it is due to product and geography mix" (Dr Lal Q2 FY25 concall). Thyrocare, reporting per test, saw realisation move on mix too — asked about a 4% rise, the CEO said plainly: "It's mix. We haven't raised prices, so it's only mix" (Thyrocare concall).

The lesson is in the two CEOs saying the same sentence. In Indian diagnostics, realisation grows through mix — more tests per patient, more wellness packages, more high-value molecular and radiology work — far more than through raw price hikes, because price is where the competition is most brutal. So when you see realisation climbing, the right follow-up is "from mix or from price?" Mix-led realisation is durable. Price-led realisation in a price war is a number waiting to reverse.

B2C vs B2B mix — the quality-of-revenue dial

This is the single most clarifying cut in the sector, and it is buried in the deck. B2C means the patient is the customer — they walk into the chain's own centre, pay retail, and the chain keeps the full margin. B2B means another lab or hospital is the customer, buying tests wholesale; volumes are huge but realisation and margin are thin. That is exactly why our two opening numbers were 26x apart: Dr Lal runs at roughly 75% B2C (Inve data, Q4 FY26) and Vijaya Diagnostic at about 92% B2C (Vijaya Q3 FY26 concall), so they capture retail rupees per patient; Thyrocare is, in its own words, "the B2B partner of choice" (Thyrocare concall), selling tests wholesale to a network of franchisee labs — hence ₹36 a test.

Neither model is "better" in the abstract; they are different businesses wearing the same coat. But the mix tells you what a margin number means and how defensible it is. A B2C-heavy chain earns its margin from brand and proximity to the patient; a B2B-heavy one earns it from scale and low-cost processing, and lives or dies by franchisee economics. Watch the direction: a B2C chain quietly leaning more on B2B to chase volume is trading margin for growth, and the blended realisation will tell on it before the commentary does.

The network — labs and collection centres

The network is the funnel that feeds the fixed plant: a relatively small number of high-cost labs, fed by a large, cheap web of collection centres (and franchisees). Dr Lal ran roughly 298 clinical labs and about 6,607 collection centres as of FY25 (Dr Lal FY25 investor presentation) and guided to 600–800 new collection centres in the year (Dr Lal Q2 FY26 concall). Thyrocare ran 39 labs in India plus one in Tanzania, fed by 10,300 active franchisees (Thyrocare Q3 FY26 concall). Vijaya runs a hub-and-spoke model — a few high-capability "hubs" surrounded by "spokes," guiding to a spoke breakeven of about two quarters and maturity around 18 months (Vijaya Q3 FY26 concall).

What "good" looks like is not the raw count — it is density and utilisation. A new collection centre or spoke is a loss-maker until enough samples flow through it; Vijaya's own guidance that the new hubs dragged EBITDA margin by ~1.1% while they ramped (Inve data, Q1 FY26) is the honest version of this. So the question for any expansion story is whether the new nodes are maturing — climbing toward the network average — or just multiplying. A chain adding centres faster than it can fill them is buying revenue and selling margin.

EBITDA margin — where it all lands, and why the level is a clue

EBITDA margin is where volume, realisation and mix finally resolve into profit, and it is the one metric you can read straight off the financials — but the level only makes sense once you know the model. Watch the spread: Dr Lal runs around 27% (Inve data, Q4 FY26), Thyrocare's consolidated normalised margin around 32% in Q3 FY26 (Thyrocare Q3 FY26 concall), and Vijaya a striking ~42% (EBITDA margin Q3 FY26 41.9% — Inve data). A homely way to hold it: a pathology lab is a kitchen, and radiology is the bar — the food keeps the lights on, but the drinks are where the margin lives. Vijaya's margin sits where it does largely because radiology (MRI, CT — high realisation, high fixed cost, high incremental margin once utilised) is a big part of its plate; a pure-pathology chain can't print that number however well it's run.

So don't reward or punish a margin against a single sector "benchmark." Read it against the company's own trajectory and its mix. A rising margin on rising volume and steady realisation is the operating-leverage flywheel working as designed. A margin propped up by deferring collection-centre expansion, or one diluted by an acquisition you weren't told would dilute it, is a different story the headline won't tell you.

How to read valuation here

Diagnostics is the rare consumer-facing healthcare business that throws off cash, carries little debt and reinvests at high returns — which is exactly why the market prices it richly and why the multiple, not the business, is usually where investors get hurt. The two lenses that fit are EV/EBITDA (it neutralises the differing depreciation and lease-accounting of asset-heavy radiology versus asset-lighter pathology, and the cash on the balance sheet) and P/E (clean comparison across the listed names). Both routinely sit well above the broader market, on the logic that volume compounds and operating leverage does the rest.

The discipline is to anchor the multiple to durable growth, and to separate it ruthlessly into volume × realisation × mix. A chain trading at a premium multiple justified by 20% revenue growth is a very different proposition if that growth is 15% volume + 5% mix (durable, fixed-plant leverage) versus 5% volume + 8% price + acquisitions (fragile, price-war-exposed, integration-risked). The second-order risk the multiple rarely prices is network density and price competition: online aggregators and well-funded rivals discounting to win the same urban patient can compress realisation across the segment, and a high multiple built on the assumption that today's realisation holds is the most expensive assumption you can make. Beyond the names above, listed diagnostics chains worth screening on these same KPIs include Metropolis Healthcare, Krsnaa Diagnostics and Suraksha Diagnostic. Cross-check the operating KPIs across peers in the KPI Screener before you accept any premium as earned.

A worked case: Dr Lal's "no price increase" strategy

Take one named company and watch said-vs-did across a year. (Illustration, not a view on the stock; a read on how management communicated through one period, not a lifetime verdict.)

In the Q2 FY25 call, with competitors raising prices, Dr Lal's then-MD was asked why his company wasn't following. His answer was a strategy stated out loud: "We are not averse to taking price increase… [but] our effort is to go down to Tier-3, Tier-4 towns. [If] we continue to take price increases, then we do not achieve that objective" (Dr Lal Q2 FY25 concall). The bet: hold price, win lower-tier-town volume, and let mix — bundled wellness panels (Swasthfit), premium tests — do the realisation lifting instead.

Now the did. Across the next several quarters revenue per patient still rose every quarter — ₹861, then ₹887, ₹880, ₹889, ₹927, to ₹956 (Inve data, Q3 FY25 through Q4 FY26) — and management repeatedly attributed it to mix, not price. Meanwhile the EBITDA margin held around 27% (Inve data, Q4 FY26), and sample volume grew 12.9% in Q4 FY26 (Inve data) even as patient volume wobbled with the fever season. Said and did, lining up: a management that guided to a volume-led, mix-led model and then produced volume-led, mix-led numbers — the realisation rose without the price hike they explicitly declined to take.

That is the kind of forward statement worth pinning to the quarter it was made and checking against what followed — the job Inve's Promise Tracker does across a portfolio so you don't re-read every transcript by hand. The point of the exercise is not to grade this one call; it is that a strategy stated and then visible in the realisation-and-mix numbers four quarters running is far more trustworthy than a margin target asserted and never reconciled.

Red flags specific to diagnostics

• Realisation rising on price, not mix — into a price war. When a chain leans on price hikes for realisation while online aggregators and discounters undercut it, the number is borrowed from next year. The durable version is mix-led, as both Dr Lal and Thyrocare's CEOs spelled out.
• Network expansion outrunning utilisation. Centres or spokes added faster than samples to fill them is operating leverage in reverse — the fixed cost arrives before the volume. A chain that won't quantify the margin drag from new nodes (the way Vijaya does — its new hubs dragged EBITDA margin by ~1.1% while they ramped, Inve data, Q1 FY26) is hiding the cost.
• A B2C chain quietly buying B2B volume. Chasing headline growth by adding thin-margin wholesale revenue dilutes the very thing that justified the premium multiple — watch blended realisation, not the growth rate.
• Margin "held" by deferring investment. A flattering margin in a quarter where collection-centre or lab additions mysteriously paused is profit borrowed from future volume.
• Acquisition-led growth with vague integration economics. Bolt-on labs can dilute group margin for years; if management won't tell you the acquired entity's standalone margin and the path to parity, assume the dilution is worse than the deck implies.

Frequently asked questions

The craft in diagnostics is refusing to be impressed by a margin or a growth rate until you've taken it apart. The business is a fixed-cost plant fed by a network, and its quality lives in three numbers under the headline — how many samples flowed, how much each patient paid, and where in the chain the rupee was captured. So invert the question you bring to the results. Don't ask "is this a high-margin, fast-growing business?" Ask: if this management were buying growth with discounts and unfilled centres while the price war eats realisation, what would the numbers look like — and does the volume-realisation-mix split rule that out?

Where this lens can be wrong. The strongest case against everything above is that the most important variable in Indian diagnostics may be one no KPI captures: distribution-led disruption. A well-funded online aggregator can subsidise tests to acquire patients at a loss, resetting the realisation the whole sector can charge — and a chain with pristine volume, realisation and mix today can still see that realisation compressed by a competitor who doesn't need to make money on the test. Reading the operating numbers tells you which incumbent is best-run; it does not tell you whether the rules of the game are about to change. And a single fever season, as Dr Lal's own 2.7% volume quarter showed, can make a quarter's volume look better or worse than the underlying trend. The honest claim is narrower than it looks: these metrics tell you whether this management's machine is getting more efficient — not whether the market it operates in will still pay the same price three years out.

And the owner's question to sit with before buying any diagnostics company: what must I believe about realisation five years out — not this quarter's volume — for this chain to still be compounding? If the answer leans on price increases the company has explicitly said it won't take, or on a network that hasn't yet filled, you've read the headline, not the business.

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