How a Pharma Business Earns Money: Brands vs Generics

In May 2026, Cipla told its analysts about a factory it doesn't even own. A partner in Rodopi, Greece — a company called Pharmathen — makes one injection, Lanreotide, that Cipla sells in America. The US drug regulator walked in, didn't like what it saw, and the line stopped. Production halted. The product vanished from Cipla's US shelves until the regulator comes back to re-check (Angel One, Feb 2026).

One inspection, in a country Cipla's customers will never visit, took a product off the market. If you want to understand a pharma business, start there — because that is the whole business in one sentence.

Think of a pharma company as a kitchen

Picture a restaurant kitchen that has built two very different businesses out of the same stoves.

The first is a neighbourhood mess hall it has run for decades. Regulars come every day, know the cook by name, pay a fair price, and rarely switch. Prices drift up a little each year. It's not exciting. It's dependable.

The second is a giant canteen contract that feeds a faraway corporate campus. The pay is huge when you win it. But the campus runs a brutal auction — three other caterers bid on the exact same thali, and whoever charges less wins. And the campus has one non-negotiable rule: a health inspector can show up unannounced, and if your kitchen fails, you lose the contract that day. Not after a warning. That day.

A pharma company is that kitchen. The mess hall is its domestic branded business. The campus contract is its US generics business. Hold both pictures in your head and almost everything about the financials starts to make sense.

The mess hall: domestic brands earn slowly and stick

A branded medicine is one sold under a company's own name to Indian doctors and chemists — think of the inhaler your family always buys by brand, not by chemical name. Doctors prescribe what they trust, patients refill what works, and nobody switches over a few rupees. That loyalty is the moat (the durable advantage a business defends, the idea we covered earlier in this series).

For Cipla, this is the steady half. In FY25 its "One India" business earned about ₹11,610 crore — roughly 42% of total revenue (Cipla FY25 results). On its FY26 call, management guided to "strong double-digit growth as well as a market beating growth" in India for the next two years (Inve data, 2026) — the unglamorous, compounding kind of growth a mess hall throws off year after year.

This is the half that lets the whole company sleep at night. A relationship with a doctor doesn't disappear because someone in another time zone underbid you by 4%.

The campus contract: US generics pay big and bleed

A generic is a copy of a drug whose patent has expired — same molecule, no brand, sold to whoever wants it cheapest. India is extraordinarily good at this. The country accounts for about 20% of global generic-drug exports by volume, which is why it's called the "pharmacy of the world" (IBEF). In the United States, where more than nine in ten prescriptions are generic, Indian firms make a large share of the pills Americans actually swallow.

The catch is the auction. When several companies sell the identical generic, the only lever left is price — and it only goes down. The industry calls this price erosion: the same product earns a little less every year as rivals pile in. As one third-party analyst summary put it, "the US generics market is highly competitive, leading to constant price erosion" (Whalesbook analyst commentary, 2026 — third-party, not a company filing).

For Cipla, North America was 29% of FY25 revenue — its highest-ever US$934 million (Cipla FY25 results). The way you survive the auction is to keep cooking dishes nobody else can — complex generics, inhalers, peptides — so you're not in a four-way price war on a plain thali. That is what R&D buys.

R&D is the kitchen's recipe budget

To keep selling things rivals can't copy, a pharma company spends heavily on developing and filing new products. Cipla guided to about ₹2,000 crore of R&D in FY26, and explained the rising cost per filing by moving into "more respiratory, more peptide and more Oligo" — harder recipes, fewer of them, each worth more (Inve data, 2026).

Here's the trap for a new investor: that spending hits profit today, but the payoff arrives years later — if it arrives. Watch what it did to the numbers. Cipla's net profit was ₹4,153 crore in FY24, ₹5,269 crore in FY25 — and then about ₹3,862 crore over the most recent twelve months, even though sales barely moved (Inve data, 2026). Operating margin fell from roughly 26% to 21% across that stretch. Management was blunt about why: "we have made a lot of investment in the last 1 or 2 years, both on people as well as on R&D … that investment phase is coming to an end" (Inve data, 2026).

Read that the way an owner would. The kitchen is paying chefs and testing recipes now for dishes it hopes to sell in FY27 and FY28. Whether that was money well spent is the entire question — and you only get the answer with a two-to-three-year lag.

The surprise inspection that can lose you the contract

Now the rule that makes pharma unlike almost any other business. To sell a drug in America, the factory that makes it must pass inspection by the US Food and Drug Administration (USFDA). These visits are often unannounced, and the regulator grades the kitchen in three words.

The FDA itself defines them plainly. NAI (No Action Indicated) means "no objectionable conditions or practices were found." VAI (Voluntary Action Indicated) means problems were found but the agency won't act — fix them yourself. And OAI (Official Action Indicated) means the facility is in an unacceptable state of compliance, and that "regulatory and/or administrative actions are recommended" (FDA, Inspection Classifications).

OAI is the failed hygiene inspection. It can freeze new approvals from that plant and halt supply — exactly what happened to Cipla's Lanreotide when its Greek partner's facility was classified OAI after nine inspection observations on "contamination control, aseptic handling and data management" (Angel One, Feb 2026). On the call, management could only say it expects "a reinspection from the FDA" and "closer visibility" next quarter (Inve data, 2026). A product that earns real money, sitting idle, waiting for an inspector to fly back.

This is why a single line in a results announcement — "we received an OAI for our X facility" — can knock a pharma stock down hard. It isn't a bad quarter. It's the contract suspended until further notice. To be fair, most facilities do remediate and re-qualify, so an OAI is usually a suspension-and-repricing event rather than a permanent loss — the supply resumes once the inspector signs off. The reason it still rattles the stock is the open-ended "until further notice": the market can't price how long the kitchen stays dark.

Reading the kitchen's books

So when you open a pharma company's numbers, four questions tell you most of what matters. Cipla is a useful worked example — not a recommendation to buy or sell it, just a clear specimen.

• How much comes from steady brands vs the auction? Cipla: ~42% India, ~29% US (FY25). The bigger and faster-growing the brand half, the calmer the earnings.
• Is the US business defending itself with complex products? Watch R&D as a share of sales and what it's spent on. Plain generics get arbitraged to nothing; complex ones don't.
• Is the balance sheet strong enough to survive a stoppage? A factory can go dark for a year. Cipla carries almost no debt — borrowings are a rounding error against ₹34,000+ crore of reserves (Inve data, 2026) — so it can fund remediation and wait. A leveraged peer in the same spot is in real trouble. But a clean balance sheet only buys time to survive a stoppage; it does not cure revenue concentrated in a single plant, product, or partner. The Lanreotide loss is the proof — all that cash did not stop the product coming off the market.
• What's the inspection record, and what did management actually say about it? This is the one most beginners skip — and it's the one that moves the stock.

That last question is the hardest to do by hand, because the evidence is buried in earnings-call transcripts, quarter after quarter, across every stock you own. Did the company say "resupply in H1" last year and quietly push it out? (Checking that systematically is its own skill.) Cipla's Lanreotide resupply guidance, for instance, shows as delayed in our tracking, and its "closer to $1 billion" US guidance was diluted to a year-end run-rate rather than a full-year number (Inve data, 2026). Following those threads across a whole portfolio is exactly what Promise Tracker is for — it reads the calls so you can watch what management does versus what it said.

Where this view could be wrong

The honest counter-argument: not every pharma company is shaped like this kitchen. A pure domestic player barely faces the auction or the inspector. And a strong US franchise in genuinely hard-to-make products can out-earn a sleepy domestic brand for years. The two-kitchen picture is the right starting lens for the big diversified Indian names — Cipla, Sun, Dr Reddy's, Lupin — not a law that fits every ticker. Always check which kitchen the company you're holding actually runs.

Frequently asked questions

Related reading

• How a Bank Makes Money: CASA, NIM and Bad Loans
• Order Book: How a Capital-Goods Business Earns
• How a Cement Company Makes Money in India