Benchmarks that matter: R&D cost, success rates and time to market, MBA Training, MBA Training
3/4+150 XP
Benchmarks that matter: R&D cost, success rates and time to market
# Benchmarks that matter: R&D cost, success rates and time to market
A single number, $2.6 billion, has shaped pharma strategy conversations for over a decade. It comes from a 2014 Tufts Center for the Study of Drug Development study estimating the fully-loaded cost to bring one new drug to market, including the cost of failures. Critics call it industry-funded and inflated. Defenders say it's the best available estimate. Both can be true. This lesson shows you where these benchmark numbers come from, how to recalculate them yourself, and how to avoid being misled by any single figure, including this one.
Why one number never tells the whole story
The $2.6B figure (Tufts, 2014, estimate) bundles two very different things:
Out-of-pocket cost: what a company actually spends on the drugs that succeed.
Capitalized cost: out-of-pocket cost plus the "opportunity cost" of capital tied up over years, plus an allocation of the cost of every drug that failed along the way.
That second part does most of the heavy lifting. Roughly half of the $2.6B is the imputed cost of capital, not cash spent. A 2020 JAMA study (Wouters, McKee, Luyten) using SEC filings of 63 companies estimated a lower median of about $1.1 billion per approved drug (estimate, 2020 data). Neither number is "wrong." They answer different questions: economic cost of capital deployed versus accounting cash spent.
Lesson: always ask "cost of what, measured how" before quoting an R&D benchmark.
The core vocabulary you need
R&D
: Research and Development. In pharma, split roughly into discovery/preclinical and clinical (human trials).
FDA: US Food and Drug Administration, the primary US regulator for drugs and biologics.
EMA: European Medicines Agency, coordinates EU-wide drug evaluation (national agencies like Germany's BfArM still matter).
NDA / BLA: New Drug Application / Biologics License Application, the FDA submission seeking approval.
IND: Investigational New Drug application, needed before human trials can start.
Phase I, II, III: Clinical trial stages testing safety (I, small, healthy or patient volunteers), efficacy/dosing (II), and confirmatory efficacy at scale (III, often hundreds to thousands of patients).
POS (Probability of Success): chance a drug entering a given phase eventually reaches approval.
NME/NCE: New Molecular Entity / New Chemical Entity, a genuinely novel drug (not a reformulation).
Patent cliff: the period when a blockbuster drug loses patent protection and faces generic or biosimilar competition.
Biosimilar: a near-copy of a biologic drug, the biologic equivalent of a generic.
Orphan drug: treats a rare disease; comes with incentives (tax credits, market exclusivity) under laws like the US Orphan Drug Act (1983).
Success rates: the number that matters most day to day
Executives and investors track clinical success rates far more than the aggregate cost figure, because it drives portfolio math.
A widely cited benchmark (BIO, Biotechnology Innovation Organization, with Amplitude and QLS Advisors, "Clinical Development Success Rates," last major edition covering 2011-2020 data) puts overall likelihood of approval from Phase I at around 7.9% across all disease areas (estimate). By phase, approximate transition rates often cited:
| Transition | Approx. success rate (estimate) |
|---|---|
| Phase I → II | ~52% |
| Phase II → III | ~28-30% |
| Phase III → Approval (NDA/BLA filing and clearance) | ~57-60% |
Oncology tends to run lower overall (large patient heterogeneity, tough endpoints); infectious disease and vaccines can run higher.
A worked calculation you'll actually use
Say a mid-cap biotech has 20 Phase I assets and wants to estimate how many will reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition → market.
20 × 0.079 (Phase I-to-approval POS) ≈ 1.6 approved drugs expected from that cohort.
This is why pharma runs large, diversified pipelines rather than betting on one molecule, and why licensing deals price early-stage assets at a steep discount to peak sales potential: the POS discount is doing most of the work in that valuation.
Time to market: the other half of the benchmark story
Cost and success rate only matter alongside time. Commonly cited estimate: total clinical development (IND to approval) averages 10 to 12 years (Tufts/DiMasi estimates, various years), of which clinical trials themselves (Phase I through III) take roughly 6 to 7 years.
FDA review timelines are more precise because they're regulatory targets, not estimates of biology:
Standard review: FDA target of 10 months from filing acceptance.
Priority review: FDA target of 6 months, granted for drugs addressing serious conditions with meaningful improvement over existing therapy.
Accelerated approval, Breakthrough Therapy, and Fast Track are additional FDA designations that can shorten development, not just review (details: FDA's expedited programs guidance).
In Europe, EMA's centralized procedure has a standard review clock of 210 active days (paused for company responses to questions), after which the European Commission grants marketing authorization valid across the EU.
Market size: the numbers to anchor conversations
As of recent estimates (IQVIA Institute reports, 2023-2024 editions), rough orders of magnitude worth knowing:
Global pharmaceutical market: approximately $1.6 to 1.7 trillion in annual spending (estimate, 2023).
US share: roughly 45-50% of global pharma spending, the single largest national market by far.
Europe (top 5: Germany, France, Italy, Spain, UK combined): a much smaller aggregate share, often cited around 20-25% of global spending combined (estimate).
US R&D spend by industry (PhRMA member companies): commonly cited at over $100 billion per year in the US alone (estimate, PhRMA annual figures).
Why the US dominates: higher list prices, faster reimbursement decisions in many cases, and a large commercially insured population. Europe's more centralized health technology assessment (HTA) processes (like Germany's IQWiG or the UK's NICE) tend to negotiate prices down faster after launch.
Knowledge check
1. Why do capitalized cost estimates for drug development tend to be substantially higher than out-of-pocket cost estimates?
2. A colleague cites a single R&D cost figure to justify a pricing decision. What is the most important follow-up question to ask before accepting the number?
3. Two studies estimate the cost of developing an approved drug and arrive at very different figures. What is the most accurate way to interpret this discrepancy?
MULTIPLE CHOICE
4. Select ALL correct answers describing what the widely cited $2.6 billion figure includes or reflects.
Select all the correct answers.
MULTIPLE CHOICE
5. Select ALL correct answers about how to responsibly use R&D benchmark figures in strategy discussions.
Select all the correct answers.
Practical due diligence: how to use these numbers without being fooled
When you see a headline pharma statistic, run this quick checklist:
1. Whose number is it? Industry association (PhRMA, EFPIA), independent academic study, or a single company's investor deck? Each has different incentives.
2. What population? "All drugs" versus "oncology only" versus "orphan drugs" changes success rates and costs by multiples.
3. Capitalized or out-of-pocket cost? As shown above, this alone can double the number.
4. What year and phase mix? Success rate benchmarks change as the pipelinepipelineAll active sales opportunities across the stages of the sales process, together with their combined potential value and probability of closing.View full definition → mix shifts (more biologics and cell/gene therapies now than a decade ago, and these have different risk profiles than small-molecule drugs).
5. Is time-to-market including or excluding preclinical discovery? "10 years" and "15 years" are both defensible depending on the starting clock.
A good habit: whenever you quote a benchmark in a memo or pitch, add the source and year in parentheses, exactly like this lesson does. It signals rigor and protects you when someone challenges the number.
For a primary source to bookmark, the FDA's Drug Approvals and Databases hub lets you check actual approval dates and review types for any drug, a fast way to sanity-check "average" timeline claims against real cases.
Key Takeaways
The famous "$2-3B per drug" figure is a capitalized cost estimate (Tufts, 2014) that includes cost of capital and failure allocation; out-of-pocket estimates run closer to $1.1B (JAMA, 2020, estimate). Always ask which definition is being used.
Clinical success rates (~7.9% Phase I to approval, BIO/QLS estimate) are the number that actually drives pipelinepipelineAll active sales opportunities across the stages of the sales process, together with their combined potential value and probability of closing.View full definition → sizing and licensing valuations; run the multiplication (assets × POS) yourself rather than trusting a single headline stat.
Time to market is roughly 10-12 years end to end (estimate), but FDA review itself has hard targets (10 months standard, 6 months priority) worth knowing precisely since they're regulatory commitments, not statistical averages.
The US is roughly half of global pharma spending (estimate, IQVIA), Europe's top five markets combined are meaningfully smaller; this asymmetry explains why launch sequencing and pricing strategy usually start in the US.
Before repeating any pharma benchmark, check the source, the population studied, the cost definition, and the year. That five-second habit prevents most public misuses of these numbers.