Pharma is shorthand for the pharmaceutical industry, the global network of companies that research, develop, manufacture, and sell medications. It spans everything from the small lab synthesizing a new chemical compound to the massive corporation distributing millions of pill bottles to pharmacies worldwide. Global prescription drug sales are projected to top $1.7 trillion by 2030, making pharma one of the largest industries on the planet.
What Pharma Companies Actually Do
At its core, a pharmaceutical company’s job is to find substances that treat or prevent disease, prove they work, get them approved by regulators, manufacture them at scale, and sell them. That description sounds simple, but each step involves years of specialized work and enormous financial risk.
Traditional pharma companies develop drugs primarily from chemical and synthetic sources. These are called small-molecule drugs: think of a standard pill like ibuprofen or a cholesterol-lowering statin. The molecules are relatively small, can usually be taken by mouth, and follow well-established manufacturing processes. Biotechnology companies, by contrast, create treatments using living organisms, such as engineered proteins, antibodies, or cell therapies. These biologics are far more complex to produce because proteins have variable structures in their folding and surface chemistry, making it difficult to maintain consistency from one manufacturing batch to the next.
In practice, the line between “pharma” and “biotech” has blurred considerably. Companies like Johnson & Johnson, Eli Lilly, and Merck are major players in both spaces. The term “biopharma” now describes companies that combine biological and chemical methods in their research. When most people say “pharma,” they’re referring to this entire ecosystem.
How a Drug Goes From Lab to Pharmacy
Bringing a new drug to market is one of the most expensive and failure-prone processes in any industry. Only about 10% of drugs that enter clinical trials ultimately win regulatory approval. The journey typically unfolds over a decade or more, moving through distinct stages.
First comes discovery, where researchers identify a biological target involved in a disease and search for compounds that interact with it. Traditional high-throughput screening, where thousands of compounds are tested rapidly, yields only about a 2.5% hit rate. Once a promising compound emerges, it goes through preclinical testing in the lab and in animal models to assess basic safety.
If a compound survives preclinical work, it enters human clinical trials in three phases. Phase I tests the drug in a small group of healthy volunteers to evaluate safety and dosing. Phase II expands to patients with the target disease to measure whether the drug actually works. Phase III involves large-scale trials, sometimes with thousands of participants, to confirm effectiveness and monitor side effects across diverse populations. Regulators generally expect results from two well-designed clinical trials before granting approval, though for rare diseases where multiple trials aren’t feasible, convincing evidence from a single trial can suffice.
A study published in JAMA Network Open estimated the average cost of developing a single new drug at roughly $173 million in 2018 dollars. That figure covers direct development costs including post-marketing studies. But most drug candidates fail, and when the cost of those failures is factored in, the average rises to about $516 million. Add in capital costs (the money tied up over years that could have been invested elsewhere) and the total climbs to roughly $879 million per successful drug. Costs vary widely by therapeutic area, ranging from about $73 million for drugs targeting the urinary system to $297 million for pain and anesthesia treatments.
How Drugs Get Approved
In the United States, the FDA’s Center for Drug Evaluation and Research reviews the evidence a company submits and makes an independent judgment: do this drug’s health benefits outweigh its known risks for the intended population? That risk-benefit calculation is context-dependent. A drug for a life-threatening cancer with no existing treatment can be approved with side effects that would be unacceptable for a drug treating a mild condition.
For serious or life-threatening diseases, the FDA offers an accelerated approval pathway. This allows a drug to be approved based on a “surrogate endpoint,” a lab measurement or physical sign that is reasonably likely to predict a real clinical benefit, even before long-term outcomes are confirmed. The trade-off is that the company must conduct follow-up studies after the drug reaches the market. If those studies fail to verify the predicted benefit, the FDA can withdraw approval.
Once approved, every drug carries an FDA-approved label describing its benefits, risks, and how to manage those risks. For drugs with particularly serious safety concerns, the manufacturer may need to implement a formal risk management plan that includes additional safeguards like restricted distribution or mandatory patient monitoring.
How Drug Pricing Works in the US
The price you pay for a medication is shaped by a chain of intermediaries, not just the company that made it. Pharmacy benefit managers (PBMs) sit at the center of this chain. The three largest, Express Scripts, CVS Caremark, and OptumRx, act as middlemen between drug manufacturers, insurance companies, and pharmacies. They build the formularies that determine which drugs your insurance covers, negotiate rebates with manufacturers, and set up pharmacy networks.
PBMs make money in several ways that aren’t always visible to patients. One common practice is spread pricing: the PBM reimburses a pharmacy one amount for a prescription (say, $30) while charging the insurance plan a higher amount ($45, for example) and keeping the difference. PBMs also negotiate rebates from manufacturers in exchange for favorable formulary placement, but they aren’t required to disclose how much of that rebate they keep versus how much they pass along to the insurer or patient. Critics argue that this lack of transparency drives up costs rather than lowering them, because PBMs can profit by favoring higher-priced drugs that generate larger rebates.
The result is a system where PBMs effectively determine which medications patients can access and afford through tiered copayment structures, prior authorization requirements, and formulary decisions.
Patents, Generics, and the Patent Cliff
When a company develops a new drug, patents grant a period of market exclusivity, typically 20 years from the date of filing, though much of that time is consumed by development and clinical trials. Once the patent expires, other manufacturers can produce generic versions. They don’t need to repeat the full slate of clinical trials. Instead, they file an abbreviated application and simply demonstrate that their version is bioequivalent, meaning it delivers the same active ingredient at the same rate and concentration in the body.
The first generic manufacturer to successfully file gets 180 days of market exclusivity before other generics can enter. For the original company, the revenue drop after patent expiration can be steep and sudden, a phenomenon the industry calls a “patent cliff.” A blockbuster drug generating billions in annual sales can lose the majority of its revenue within months as cheaper generics flood the market. This dynamic is a major reason pharma companies invest so heavily in discovering new drugs and extending their product pipelines.
How AI Is Changing Drug Discovery
Artificial intelligence is reshaping how pharma companies find and develop new drugs. The traditional process has long been plagued by low success rates and enormous costs, and AI tools are starting to compress timelines at several stages. In predictive toxicology, machine learning models analyze molecular features to flag safety problems earlier, reducing the number of candidates that fail expensive late-stage trials. One approach using graph-based neural networks to model how drugs interact with tissues reduced simulation errors by 30% compared to older computational methods.
AI is also speeding up clinical trials themselves. Tools that analyze electronic health records can optimize which patients are enrolled in a trial, reducing required sample sizes by 25 to 40% without sacrificing statistical power. In oncology trials, machine learning-driven patient matching has led to 35% faster enrollment. These improvements don’t eliminate the fundamental challenges of drug development, but they chip away at the time, cost, and waste at nearly every step.