Vose SoftwareModelRisk example · Pharmaceuticals and biotech
Launch Sales Forecast: Peak Sales and Uptake
How big will the new drug be? Peak sales from the patient funnel, the uptake curve and the competition.
From the model’s saved run
Peak sales
Net sales in the best year, US$m.
| Measure | Unit | Value |
|---|---|---|
| Base case | US$m | 924 |
| Mean | US$m | 1,064 |
| 10th percentile | US$m | 533 |
| Median | US$m | 957 |
| 90th percentile | US$m | 1,714 |
| Chance of a blockbuster (above US$1 billion) | probability | 46% |
| Chance of falling short of the base case | probability | 47% |
| Year of peak sales after launch, median | year | 6 |
Histogram of Peak sales from the saved simulation.
The business question
A company will launch a new oral treatment for moderate-to-severe rheumatoid arthritis in the US and the five largest European markets in 2028. About four million adults there have the disease. About 70% are diagnosed and treated, and about a quarter of those have moderate-to-severe disease that has failed first-line treatment — the drug's label.
The commercial team has given, for each year after launch, the market share it is 90% sure to beat and the share it is 90% sure not to exceed. A drug that takes off does so in every year. Patients stay on treatment about eight and a half months a year, at a net price of about US$17,000 a year, falling 2% a year.
A competitor in the same class has a 60% chance of launching three to six years after us and would take about a quarter of our share. The base case, built on the most likely values and the middle of the share range, gives peak sales of about US$920 million.
- Who it is for
- Commercial forecasting and brand teams, portfolio and business-development teams, finance.
- Main outputs
- Peak sales: base case, mean, P10, P90 and the chance of a blockbuster · the year of peak · sales year by year (fan chart) against the base case · cumulative sales · what drives peak sales.
How the model works
Three sheets after this one, read from left to right. Each one feeds the next.
Inputs sheet
The patients with the disease, the share treated and the share eligible (PERT). The market share the commercial team expects in each year after launch, as a P10 and a P90, and how strongly the years move together; months on treatment, net price and its erosion. The competitor's chance, entry years and impact.
Forecast sheet
VoseTimeSME2Perc draws ten years of market share: in each year a Lognormal through the P10 and the P90, correlated across the years. Patients treated = eligible patients × share × the competitor's effect. Sales = patients × months on treatment × net price. The peak and its year are found in every iteration.
The base case repeats the calculation with the most likely values and the middle of the share range, without the competitor.
Results sheet
Peak sales against the base case, the chance of a blockbuster and of falling short; sales by year; the inputs that drive peak sales.
ModelRisk techniques
Find similar models
- Product
- Type
- Application
- Uses
The results in detail
Sales by year
Net sales from launch, US$m: the simulated range against the base case.
Line chart: 90th percentile, Median, Mean, 10th percentile, Base case.
| Measure | Unit | Value |
|---|---|---|
| Net sales over ten years, base case | US$m | 7,491 |
| Net sales over ten years, mean | US$m | 7,485 |
| …10th percentile | US$m | 3,797 |
| …90th percentile | US$m | 11,968 |
What drives peak sales
Lowest and highest mean peak sales over ten equal groups of each input (VoseSimSensitivityMean), drawn from the overall mean.
Tornado chart: the inputs ranked by how far the mean output moves across each input’s range.
The ranking as a table 8 inputs
| Input (ranked by influence) | Lowest mean | Highest mean | Rank |
|---|---|---|---|
| Market share year 5 | 509 | 1,962 | 1 |
| Eligible share | 869 | 1,277 | 2 |
| Net price at launch | 938 | 1,204 | 3 |
| Months on treatment | 921 | 1,184 | 4 |
| Competitor launches | 964 | 1,215 | 5 |
| Diagnosed and treated | 954 | 1,165 | 6 |
| Adults with the disease | 992 | 1,130 | 7 |
| Share the competitor takes | 1,004 | 1,122 | 8 |
How to read these results
- The base case — the most likely funnel and price, the middle of each year's share range, no competitor — gives peak sales of about US$920 million. The simulation agrees on the middle (median about US$960 million) but not on the range: one launch in ten peaks below about US$530 million, one in ten above US$1.7 billion.
- The chance of a blockbuster, peak sales above US$1 billion, is about 46%. The chance of falling short of the base case is about the same, 47%. A single number — 'peak sales of US$920 million' — hides both.
Market share drives most of the range: mean peak sales run from about US$510 million in the tenth of futures with the lowest year-5 share to about US$1.96 billion in the tenth with the highest. The commercial team's uptake estimates matter far more than the epidemiology, of which only the eligible share matters much.
Analogues, market research and the first months of sales narrow the forecast most.
- The competitor matters about as much as the price and the months on treatment: it launches in 60% of futures and takes about a quarter of our share from its launch year, which the base case ignores. The ten-year total still averages about the base case, because a share above the middle of the expert's range adds more than a share below it takes away.
- The correlation between years matters too. With 0.9 a drug that starts well stays ahead. Set it lower on the Inputs sheet and good and bad years average out, the range narrows, and the forecast understates the risk.
See it move. Open the workbook, change an input on the Inputs sheet and run the simulation: every table and chart on this page updates in Excel.
Inputs and assumptions
Every assumption in the model, with its uncertainty and where it comes from. The figures are illustrative.
Patients 3 inputs
Adults with rheumatoid arthritis in the US and the five largest European markets.
| Assumption | Unit | Uncertainty | Low | Most likely | High |
|---|---|---|---|---|---|
| Adults with the diseaseEpidemiology studies: 0.5–0.8% of adults. | million | PERT | 3.60 | 4.00 | 4.40 |
| Diagnosed and treated | of patients | PERT | 60% | 70% | 80% |
| Moderate to severe, after first-line treatmentThe drug's label. | of those treated | PERT | 20% | 25% | 35% |
Treatment and price 3 inputs
| Assumption | Unit | Uncertainty | Low | Most likely | High |
|---|---|---|---|---|---|
| Months on treatment per patient a yearPatients start and stop during the year; some pause. | months | PERT | 7.0 | 8.5 | 10.0 |
| Net price per patient a year, at launchAfter discounts and rebates; US and Europe blended. | US$k | PERT | 14.0 | 17.0 | 20.0 |
| Change in net price each year | per year | Fixed | −2.0% | ||
Competitor 4 inputs
A second drug in the same class.
| Assumption | Unit | Uncertainty | Low | Most likely | High |
|---|---|---|---|---|---|
| Chance the competitor launches | probability | Fixed | 60% | ||
| Launches in year … at the earliest | year after our launch | Fixed | 3 | ||
| … at the latestEvery year in between equally likely. | year after our launch | Fixed | 6 | ||
| Share it takes from us | of our share | PERT | 15% | 25% | 40% |
Launch and market share 2 inputs
The year of launch, and how strongly the years move together: a drug that takes off does so in every year.
| Assumption | Unit | Value |
|---|---|---|
| Launch yearYear 1 of the forecast: sales by calendar year from the launch. | year | 2028 |
| Correlation between yearsThe CorrelationFactor of VoseTimeSME2Perc. | 0 – 1 | 0.90 |
Market share by year after launch 10 inputs
The commercial team's estimates: 10% chance of a share below the P10, 10% above the P90. Value in model: this iteration's share.
| Assumption | Unit | Uncertainty | P10 | P90 |
|---|---|---|---|---|
| Market share, year 1 (2028)Analogues: the last two oral launches in this class. | of eligible patients | Expert range · SME2Perc | 1.0% | 4.0% |
| Market share, year 2 (2029) | of eligible patients | Expert range · SME2Perc | 3.0% | 9.0% |
| Market share, year 3 (2030) | of eligible patients | Expert range · SME2Perc | 5.0% | 14.0% |
| Market share, year 4 (2031) | of eligible patients | Expert range · SME2Perc | 6.5% | 18.0% |
| Market share, year 5 (2032) | of eligible patients | Expert range · SME2Perc | 7.0% | 20.0% |
| Market share, year 6 (2033)Plateau: the drug's established share. | of eligible patients | Expert range · SME2Perc | 7.0% | 21.0% |
| Market share, year 7 (2034) | of eligible patients | Expert range · SME2Perc | 7.0% | 21.0% |
| Market share, year 8 (2035) | of eligible patients | Expert range · SME2Perc | 7.0% | 21.0% |
| Market share, year 9 (2036) | of eligible patients | Expert range · SME2Perc | 7.0% | 21.0% |
| Market share, year 10 (2037) | of eligible patients | Expert range · SME2Perc | 7.0% | 21.0% |
Units and conventions
- Money
- US$ million (US$m) of net sales; prices in US$ thousand (US$k) per patient a year.
- Patients
- Thousand patients treated on average in the year.
- Years
- Year 1 is the launch year on the Inputs sheet: 2028 as shipped.
- Blockbuster
- A drug with peak sales above US$1 billion a year.
- Base case
- Most likely funnel, treatment and price; the middle of each year's share range (the geometric mean of the P10 and the P90); no competitor.
Key assumptions and limitations
- Funnel
- Prevalence, treatment and eligibility constant over the ten years.
- Share
- The expert's P10 and P90 are those of a Lognormal in each year; a common factor correlates the years.
- Competitor
- Enters at the start of a year, equally likely to be year 3, 4, 5 or 6. Takes a fixed fraction of our share from then on.
- Data
- The company, the drug and all figures are invented for this example.
How to use it
- Enter your patient funnel and the market shares your team expects, as a P10 and a P90 for each year.
- Enter time on treatment, net price and the competitive threat.
- Run a simulation from the ModelRisk ribbon (10,000 samples with a fixed seed are set).
- Report peak sales as a range, not a number; spend forecasting effort on the top drivers.
The tornado chart fills in after a simulation with ModelRisk’s classic engine.
ModelRisk functions used
- VoseTimeSME2Perc
- Ten years of market share from an expert's P10 and P90 for each year, correlated across years (array).
- VosePERT
- The funnel, months on treatment, net price and the competitor's impact.
- VoseBernoulli, VoseStepUniform
- Whether the competitor enters, and in which year.
- VoseInput, VoseOutput
- Mark the uncertain inputs and the outputs for the tornado and the statistics.
- VoseSimMean, VoseSimPercentile, VoseSimProbability
- Statistics of the simulation on the Results sheet.
- VoseSimSensitivityMean
- The tornado chart: every VoseInput ranked by how far the mean peak sales move across its range.
Further reading
- Sensitivity analysis in Excel: a tornado chart step by step (Vose guide)
- Cook, A. G. (2015). Forecasting for the Pharmaceutical Industry, 2nd edition. Gower.
- Vose, D. (2008). Risk Analysis: A Quantitative Guide, 3rd edition. Wiley — eliciting expert opinion.
Download the workbook and run it yourself. It opens in Excel with ModelRisk. The free trial runs every example model.
Related example models
- ModelRisk★★★★★Bayesian Estimate of a Trial's Success ProbabilityHalfway through a Phase II trial: how likely is it to succeed — and should it stop, or carry on?A Beta prior updated with the interim data: the predictive chance that the final analysis succeeds, simulated and exact.
- ModelRisk★★★★★Drug Development Risk-Adjusted NPVWhat is a compound in Phase II worth — and should we develop it ourselves or license it out?Phases with uncertain cost, time and success, then launch and patent expiry: the textbook rNPV against the simulated NPV.
- ModelRisk★★★★★Patient Recruitment and Trial TimelinesWhen will the last patient be enrolled — and how many sites does it take to be confident of the deadline?Sixty sites, each with its own opening time and recruitment rate: a Poisson–Gamma model of enrolment, month by month.
An illustrative example: the figures are invented for teaching and do not describe a real company, project or asset. Last updated 9 October 2026.