A governance-first decision support framework for pharmaceutical launch planning
This project presents a decision support framework for pre-launch planning that quantifies how uncertainty in uptake timing and analog selection translates into revenue risk and forecast trade-offs.
The framework is designed for commercial launch planning contexts where leaders must reason under uncertainty, evaluate assumptions transparently, and communicate trade-offs clearly across functions.
Intended Use Context
- Decision Window: Pre-launch planning (12–18 months from approval)
- Domain: Oncology therapeutics (solid tumors and hematologic malignancies)
- Typical Users: Commercial, brand, and planning leaders involved in launch readiness and forecasting discussions
🔗 Interactive Shiny dashboard:
Launch Curve Forecaster (ShinyApps.io)
What is the revenue at risk if launch timing shifts under uncertainty?
This framework supports that question by:
- Structuring analog-based launch scenarios using a governance-first methodology
- Quantifying Year 1 revenue at risk under configurable launch delay assumptions
- Supporting P25 / P50 / P75 forecast trade-off discussions across stakeholders
- Documenting assumptions, exclusions, and sensitivities transparently for audit and review
| Capability | Implementation |
|---|---|
| Governance-first analog selection | Inclusion criteria locked before examining outcomes |
| Revenue-at-risk quantification | Year 1 impact from configurable delay scenarios |
| Sensitivity testing | Option to exclude the top-performing analog (by peak revenue) |
| Dynamic selection rationale | Live statistics: solid/heme count, approval-year range, median peak |
| Transparent methodology | Assumptions, limitations, and exclusions clearly documented |
✅ Structures analog-based launch scenario analysis
✅ Quantifies revenue ranges from analog variation (P25 / P50 / P75)
✅ Supports structured trade-off conversations under uncertainty
✅ Documents assumptions and limitations transparently
✅ Provides sensitivity checks to test analytical robustness
❌ Predict whether a specific launch will succeed
❌ Recommend specific investment or resourcing levels
❌ Model competitive response or market access dynamics
❌ Estimate NPV or financial returns
❌ Replace clinical, regulatory, or strategic judgment
By design:
NPV and market access modeling are excluded because they require company-specific inputs that would compromise generalizability. Revenue scenarios are treated as foundational inputs, not final financial outputs.
High-level summary with key metrics (analog count, peak revenue range, Year 1 revenue at risk), scenario comparison chart, and decision framing.
Configure analog selection criteria, test delay scenarios, and examine dynamic selection rationale with live statistics and sensitivity checks.
Examine individual analog launch curves and performance characteristics, with guardrails against over-interpretation of any single analog.
Transparent documentation of methodology, data sources, assumptions, limitations, and explicitly excluded modeling domains.
Inclusion criteria are specified before examining outcome data to ensure defensible analysis:
- Therapeutic area: Oncology (solid tumors and/or hematology)
- Peak revenue threshold: Configurable minimum (filters speculative analogs)
- Data availability: Product-level revenue disclosed in SEC filings
- Sensitivity check: Option to exclude the top-performing analog to test robustness
Percentile bands are derived from observed analog performance:
- P50 (Base Case): Median analog trajectory
- P75 (Upper Range): Upper quartile performance
- P25 (Lower Range): Lower quartile performance
Important: These represent observed analog ranges, not statistical confidence intervals.
Revenue at Risk (Year 1) = delay_quarters × (median Year 1 revenue / 4)
This is a Year 1 average-quarter approximation, explicitly labeled to match the calculation methodology.
- Demand deferral: Delayed demand is deferred, not destroyed
- Label expansion bias: Some analogs experienced post-launch indication expansion, creating potential upward bias
- Era effects: Launch dynamics differ between earlier (2013–2015) and more recent oncology approvals
| Source | Purpose | Access |
|---|---|---|
| SEC EDGAR filings | Product-level revenue (10-K reports) | SEC EDGAR |
| FDA Drug Approvals | Approval dates (T=0), indications | FDA.gov |
| Published literature | Launch curve patterns and benchmarks | Peer-reviewed sources |
All data used is publicly available. No proprietary or confidential information is included.
The framework includes established oncology launches with disclosed product-level revenue:
- Solid tumors: Keytruda, Opdivo, Ibrance, Tagrisso, Lynparza, Tecentriq
- Hematology: Imbruvica, Darzalex, Venclexta, Calquence
| Component | Technology |
|---|---|
| Language | R (4.3+) |
| Framework | Shiny (modular architecture) |
| UI components | shiny.semantic (Appsilon) |
| Visualization | ggplot2, ggiraph |
| Tables | reactable, reactablefmtr |
| Deployment | shinyapps.io |
- Governance-first thinking — lock criteria before examining outcomes
- Transparency over performance — document assumptions and limitations
- Appropriate restraint — clearly state what the tool does not do
- Decision support, not automation — inform judgment, do not prescribe
- Executive communication — lead with insights, support with data
- Revenue data from SEC filings may include ex-US sales
- Label expansion can introduce upward bias in peak revenues
- Earlier analogs may not fully reflect current launch environments
- Revenue-at-risk calculation is simplified (not a full DCF model)
- Competitive dynamics and market access are not modeled
Steven Ponce
Data Analyst | R Shiny Developer | Pharmaceutical Analytics
- 🔗 Portfolio: stevenponce.netlify.app
- 🐙 GitHub: @poncest
- 💼 LinkedIn: stevenponce
Released under the MIT License.
This project is a portfolio case study intended to demonstrate analytical framing, governance, and decision-support design using publicly available data.
It is not intended for real-world commercial decision-making and is not affiliated with any pharmaceutical company. Product names are used for analytical illustration only.
Version: 1.0.0
Last updated: January 2026