Here’s a number that should make any launch team uncomfortable: more than half of all pharma launches miss their commercial targets. Not marginally. Not by a rounding error. By enough that peak sales projections get quietly revised, mid-launch strategy reviews get called, and someone in a boardroom starts asking questions that nobody has clean answers to. Evaluate Pharma’s World Preview 2023 makes this painfully clear — underperformance at launch is the statistical norm in this industry, not the outlier.
And yet the post-mortem conversation almost always goes the same way. Execution gets picked apart. Messaging gets questioned. The sales force deployment gets scrutinised. What almost never gets examined — honestly, at the right level of rigour, with enough candour to be useful — is whether the evidence the launch was built on was actually any good.
That’s the conversation this post is trying to have.
No company names here, no product callouts. But a clear-eyed look at the pattern of decisions and evidence gaps that produce the same outcome, again and again: a launch that underperforms because the commercial intelligence behind it was wrong, incomplete, or arrived too late to change anything that mattered.
If you’re 12 to 36 months from a Continental European launch, this is the post-mortem you read now. Not after the first quarterly review tells you the uptake curve doesn’t match the model.
The Anatomy of a Failed Launch: Three Root Causes That Keep Showing Up
Failed launches are almost never the result of one catastrophic decision. What actually happens is a quieter, more insidious process — a series of evidence gaps, each one seemingly manageable on its own, that compound into something commercially significant by the time the launch lands. McKinsey’s work on what separates winning launches from the rest identified consistent structural patterns across underperforming programmes. Three root causes show up with uncomfortable regularity.
Root Cause 1: The Market Was Sized on the Wrong Numbers
The patient population was overestimated. Or the addressable market was misunderstood. Or competitive displacement was modelled on what analysts projected rather than what prescribers were actually doing.
Now here’s the thing — this almost always traces back to a single methodological mistake. Forecast models get built on total disease prevalence figures. Which sounds reasonable until you realise that total prevalence and the population your drug can actually reach are very different numbers. Run the filters — diagnosed rate, treatment eligibility, formulary access, realistic physician reach across your priority markets — and the addressable population can be a fraction of the headline epidemiology figure.
When the market is sized wrong, everything downstream is wrong. Peak sales. Sales force sizing. Inventory. Investor commitments. The error compounds through the entire commercial model, and it announces itself only after launch — when the uptake curve tells a story the forecast didn’t predict and nobody can quite explain why. See also: Transforming Pharma Market Sizing With Real-World Insurance Claims Data.
Root Cause 2: The Patient Journey Was Assumed, Not Measured
Treatment pathways were built from advisory board workshops rather than population-level data. And I understand why that happens — advisory boards are accessible, they’re fast, and the KOLs in the room are genuinely knowledgeable. But here’s what they can’t tell you: what actually happens across the broader patient population, in the real healthcare systems your drug will launch into, at a scale that’s statistically meaningful.
The gaps that matter are the ones you can’t see without claims data. Time from first relevant diagnosis code to treatment initiation. The proportion of diagnosed patients who never get treated at all — and why. The drop-off rates at each care transition. How the journey varies, sometimes substantially, between markets in Continental Europe where healthcare system structures differ. These numbers determine your real uptake velocity. And when they’re wrong, your channel strategy is wrong, your messaging is wrong, and your targeting is wrong. The launch lands in a market reality it wasn’t designed for. Real-World Treatment Pathway Analysis Using Aggregated Claims Data gets into the specifics of how this plays out.
Root Cause 3: The Payer Evidence Wasn’t Built for the Payers Who Actually Matter
The HEOR dossier was built too late. Or on the wrong data. Or — and this one stings — it was built without a clear understanding of what specific payers in specific Continental European markets actually need to see to say yes.
This is probably the most frustrating root cause because it’s the most preventable. The data needed to build a compelling comparative effectiveness argument, a credible budget impact model, a health economic case that holds up under scrutiny — it existed. It just wasn’t commissioned at the right time, from sources with the right depth for the markets in question. And so reimbursement rejections arrive after launch, payer objections surface that could have been anticipated and addressed, and the market access team finds itself retrofitting evidence to a situation it should have shaped. HEOR Simulation and Cost Modeling with Aggregated Insurance Data covers what a rigorous pre-launch payer evidence strategy actually looks like in practice.
Deloitte’s analysis of pharmaceutical launch strategy has consistently flagged payer evidence gaps as one of the most common — and most costly — drivers of underperformance across markets.
Where the Data Gaps Actually Start: 18 to 36 Months Before Launch Day
Honestly, this is the part of the conversation that most teams don’t want to have. Because it means accepting that by the time the launch team is executing, the critical evidence decisions have already been made. The market sizing model is locked. The patient journey assumptions are baked into the commercial plan. The HEOR dossier is in front of payer committees. The window to change any of it has closed.
Most launch failures are not execution failures. They are evidence failures — and they happen 18 to 36 months before launch day, during the pre-launch window when course correction is still possible and still relatively inexpensive.
The problem typically traces back to one of three structural failures. Wrong data sources — market sizing built on syndicated secondary reports rather than claims-derived patient population analysis; patient journey maps constructed from KOL input rather than treatment pathway studies; payer evidence models built on assumptions rather than observed healthcare utilisation data. Wrong vendor choices — partnerships selected for convenience or familiarity rather than fit-for-purpose coverage, therapeutic area depth, and the analytical capability needed for the specific use cases at hand. Or, more commonly than it should be, no real-world evidence strategy at all — particularly in lean pharma and biotech organisations where RWE planning begins too late, is under-resourced relative to clinical development, and is treated as a regulatory checkbox rather than a commercial asset.
Berger et al. (2017) set out the good practice framework for real-world data studies of treatment and comparative effectiveness in the European Journal of Epidemiology — and the gap between that standard and what most pre-launch teams actually commission is, in my experience, significant. Understanding which data sources are genuinely fit for purpose requires deep market knowledge that most companies don’t have sitting internally.
What Real-World Data Would Have Changed
Let’s make this concrete. For each of the three root causes, there’s a specific data solution that — applied at the right point in the pre-launch timeline — would have changed the outcome.
On Market Sizing: Claims-Based Patient Population Quantification
Pan-European health insurance claims data — diagnostics coded to ICD-10, prescriptions classified by ATC code, procedure and billing codes — lets you build a market sizing model grounded in what actually happens in the healthcare systems you’re launching into, not what secondary research estimates from the top down.
Diagnosed prevalence derived from claims analysis is materially different from epidemiological prevalence estimates. Apply treatment eligibility filters to a claims-derived patient cohort and you get a realistic addressable population — not an analyst’s projection built on assumptions about the share of total disease burden your drug might reach. Real-world prescribing patterns across Continental European markets reveal competitive dynamics that top-down estimates consistently misrepresent.
As Sherman et al. (2016) established in the New England Journal of Medicine, real-world evidence provides insights into disease burden, treatment patterns, and outcomes that clinical trial data simply cannot generate. Market sizing is exactly the kind of problem they were describing.
On Patient Journey Mapping: Population-Level Treatment Pathway Analysis
Anonymised, aggregated claims data enables longitudinal tracking of patient cohorts from first diagnosis code through treatment initiation, therapy switching, and clinical events. That’s a fundamentally different quality of insight than what an advisory board can give you.
Time-to-treatment. Drop-off rates at every care transition. The real proportion of eligible patients who never receive treatment. The uptake velocity that reflects how quickly patients actually arrive at prescription — not the optimistic assumption baked into the forecast model. All of this is measurable. But only if you’ve commissioned data from sources with sufficient longitudinal depth and population coverage across your target markets, evaluated against quality standards like the GRACE principles for high-quality observational studies.
On Payer Evidence: Real-World Comparative Effectiveness Built for the Payers Who Decide
The evidence that national HTA bodies across Continental Europe need to see isn’t generated in Phase III. It’s generated from real-world data — comparative effectiveness analysis against current standard of care in the markets where you’re seeking reimbursement, healthcare resource utilisation evidence, budget impact modelling grounded in observed costs rather than assumed unit costs.
Garrison et al. (2007) established the foundational principles for using real-world data in coverage and payment decisions for the ISPOR Real-World Data Task Force — and those principles remain central to payer evidence strategy today. Building a HEOR dossier without this foundation isn’t a cost saving. It’s a gamble with reimbursement. And it’s a gamble that frequently doesn’t pay off.
The challenge most pharmaceutical companies face isn’t finding data providers — it’s knowing which combination of data capabilities, market coverage, and analytical depth aligns with what their specific payers in their specific markets actually require. That judgement call requires expertise that goes beyond what most internal teams can credibly claim. Expert advisors who understand both your business objectives and the vendor landscape can accelerate your path to value while helping you avoid the kind of costly mismatches that only become visible after a submission is rejected.
See also: Using Claims Data for Market Research in Pharma and MedTech.
The Pre-Launch RWE Checklist: Evidence Questions That Should Have Answers 18 Months Before Launch
Think of this as the stress test. If you can’t answer these questions with confidence — backed by real-world data, not secondary research assumptions — you have evidence gaps that will surface at some point in the launch. Better to find them now.
Market Sizing
- What is the diagnosed prevalence of the target condition across your priority Continental European markets, based on claims data rather than epidemiological estimates?
- What proportion of diagnosed patients are actually treatment-eligible under the anticipated label?
- What are real-world prescribing patterns for current treatment options — and where does your asset’s positioning realistically fit within that competitive picture?
Patient Journey
- What is the actual time from first relevant diagnosis code to treatment initiation in your therapeutic area, across your priority markets?
- What proportion of eligible patients are diagnosed but never reach treatment — and at which care transition does the drop-off occur?
- How does the patient journey vary across national markets in Continental Europe, and does your launch strategy account for those differences?
Payer Evidence
- What comparative effectiveness evidence will national HTA bodies in your priority markets require at submission — and do you know whether that differs meaningfully between markets?
- What are the real-world healthcare resource utilisation costs associated with the current standard of care in those markets?
- Can you model your asset’s budget impact using claims-derived cost data rather than assumed unit costs?
- Is your HEOR evidence strategy aligned with what specific payers need to see — or is it built to a generic European payer archetype that doesn’t quite fit any of them?
Both the FDA’s guidance on assessing real-world data for regulatory decision-making and the EMA’s DARWIN EU network make the same underlying point: the quality and fitness-for-purpose of real-world data used in regulatory and payer submissions isn’t a technical detail you sort out later. It’s a strategic decision that shapes the entire evidence trajectory from the point it’s made.
Before you start evaluating data providers, defining your requirements clearly across these dimensions is essential. The right expert input at this stage means you’re comparing options on the dimensions that actually matter for your use cases — not just on headline coverage figures or day rate.
The Cost of Getting It Wrong Versus the Cost of Getting It Right
Let’s talk about the financial asymmetry here — because it’s more extreme than most launch investment cases make explicit.
A robust pre-launch real-world evidence programme covering market sizing, patient journey, and payer evidence is a material investment. But it’s a bounded one. It happens in a defined window. It produces intelligence that’s reusable across commercial, medical, and market access functions. And — critically — it’s commissioned at the point where the insights can still change decisions that haven’t been made yet.
A missed launch target is a different category of problem entirely.
Deloitte’s ongoing analysis of pharmaceutical R&D returns tracks the declining internal rate of return on drug development investment — and the pattern is consistent. Late-stage and post-launch commercial underperformance is one of the primary drivers of that decline. Miss a peak sales forecast by 30% in year one and you won’t recover it by year three. Uptake curves that start slowly rarely accelerate to compensate for the lost ground. The revenue impact is largely permanent.
Post-launch course correction is reactive and expensive. Rebuilding a payer evidence dossier after a reimbursement rejection costs more — in elapsed time, consulting fees, and delayed revenue — than building it correctly before submission. A reforecast model that emerges from a weak first quarter doesn’t restore investor confidence. It manages the rate of its decline.
Smietana et al. (2016) in Nature Reviews Drug Discovery documented the compounding cost of late-stage failure in drug development. The logic translates directly to commercial failure: the further downstream the problem surfaces, the more expensive it is to address.
Reframing the internal conversation is important here. Pre-launch RWE investment is not a cost centre. It’s launch insurance — with a demonstrably better expected value than the alternative. The financial asymmetry between investing in the right data pre-launch versus absorbing a missed launch target and attempting to recover from it should be part of every launch investment case. Data-Driven Product Launch Planning with Aggregated Patient Data explores the commercial case in more depth.
As Kesselheim and Doherty (2021) noted in Health Affairs, the opportunities presented by real-world evidence are substantial — but realising them requires ensuring the data is genuinely fit for purpose. That judgement takes expertise and time to develop, which is precisely why starting the conversation early matters.
From Post-Mortem to Prevention
The pattern described throughout this post is not inevitable. The evidence gaps that produce failed launches are identifiable, addressable, and — with the right data strategy in place at the right point in the pre-launch timeline — entirely preventable. But prevention requires acting before the window closes, not after the first quarterly review makes the problem visible.
If your asset is 12 to 36 months from launch across Continental European markets, the questions in the checklist above should be driving active decisions right now. Not because the answer is the same for every therapeutic area or every market combination. The Pan-European health data landscape doesn’t work that way — coverage varies, analytical capabilities differ, and what is fit-for-purpose for market sizing in one context may be inadequate for patient journey analysis in another.
Makady et al. (2017) mapped the variation in how real-world data is actually used across European HTA bodies — and the picture is not uniform. What satisfies a national HTA process in one market may be insufficient in another. That variation has to be built into evidence strategy from the beginning, not accommodated retrospectively after a submission is rejected.
The ISPOR Real-World Evidence Transparency Initiative is clear on this: the guiding principles for credible, decision-grade real-world evidence start with study design and data source selection decisions made long before the first analysis is run.
Navigating the Continental European data landscape — understanding which sources are genuinely fit for purpose across markets, therapeutic areas, and payer audiences — requires expertise in both pharmaceutical commercialisation and the evolving real-world evidence ecosystem. The gap between licensing claims data and extracting actionable intelligence that actually changes a launch outcome is substantial. Bridging that gap is where expert guidance earns its value.
If you’re building your launch evidence strategy for Continental European markets and want to stress-test your current approach, we’d welcome the conversation. Explore MEDDDICAL’s commercial analytics advisory services — or contact us directly to discuss your specific evidence requirements before the pre-launch window closes.
Sources & References
- Evaluate Pharma (2023). World Preview 2023, Outlook to 2028. Evaluate Ltd. https://www.evaluate.com
- Ringel, M., et al. (2013). Breaking the risk-reward doom loop in pharmaceutical R&D. McKinsey & Company. https://www.mckinsey.com
- Grubert, N., et al. (2018). What separates winning pharmaceutical launches from the rest. McKinsey & Company. https://www.mckinsey.com
- Fortner, K. (2022). Why do so many drug launches underperform? Deloitte Insights. https://www2.deloitte.com
- Berger, M. L., et al. (2017). Good practices for real-world data studies of treatment and/or comparative effectiveness. European Journal of Epidemiology, 32(11), 945–947. https://link.springer.com
- Khozin, S., Blumenthal, G. M., & Pazdur, R. (2017). Real-world data for clinical evidence generation in oncology. Journal of the National Cancer Institute, 109(11). https://academic.oup.com/jnci
- Sherman, R. E., et al. (2016). Real-world evidence — what is it and what can it tell us? New England Journal of Medicine, 375(23), 2293–2297. https://www.nejm.org
- Garrison, L. P., et al. (2007). Using real-world data for coverage and payment decisions: the ISPOR Real-World Data Task Force report. Value in Health, 10(5), 326–335. https://www.valueinhealthjournal.com
- Dreyer, N. A., et al. (2018). GRACE principles: recognizing high-quality observational studies of comparative effectiveness. American Journal of Managed Care, 16(6), 467–471. https://www.ajmc.com
- U.S. Food and Drug Administration (2023). Real-world data: assessing electronic health records and medical claims data to support regulatory decision-making for drugs and biological products. FDA Guidance Document. https://www.fda.gov
- European Medicines Agency (2023). Data Analysis and Real-World Interrogation Network (DARWIN EU). EMA. https://www.ema.europa.eu
- Deloitte Centre for Health Solutions (2024). Measuring the return from pharmaceutical innovation. Deloitte. https://www2.deloitte.com
- Smietana, K., Siatkowski, M., & Møller, M. (2016). Trends in clinical success rates. Nature Reviews Drug Discovery, 15(6), 379–380. https://www.nature.com
- Kesselheim, A. S., & Doherty, M. (2021). Real-world evidence in drug regulation — challenges and opportunities. Health Affairs, 40(2). https://www.healthaffairs.org
- Makady, A., et al. (2017). Policies for use of real-world data in health technology assessment. Value in Health, 20(3), 520–532. https://www.valueinhealthjournal.com
- ISPOR (2022). Real-world evidence transparency initiative: guiding principles. ISPOR Special Report. https://www.ispor.org
