Biotech rNPV and sum-of-the-parts model
Three programmes, stage probabilities, launch curves, SOTP. Value a development-stage biotech programme by programme: build a launch curve from patients and price, risk-adjust by stage, discount to an rNPV, and sum the parts to a price per share.
By Surojit Chakraverti, founder of L3VLUP and an investor running a long-short healthcare and technology equities strategy.
On this subject: biotech rnpv and sum-of-the-parts model.Updated 30 September 2026
Who builds it, and for whatThe valuation behind every biotech initiation, every healthcare fund’s position and every licensing negotiation: a company that has no earnings valued on what its pipeline would earn if it worked, weighted by the chance that it does. Built by healthcare equity analysts, biotech investors and business development teams, and argued over line by line, because the answer is mostly probability and launch curve rather than discount rate.
| A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Assumptions | ||||||||||||||||||||||
| 2 | Blue cells only. $ millions except per-patient prices; patients in thousands; annual periods Y1 to Y20. The company is invented. | ||||||||||||||||||||||
| 4 | Year | Unit | Y1 | Y2 | Y3 | Y4 | Y5 | Y6 | Y7 | Y8 | Y9 | Y10 | Y11 | Y12 | Y13 | Y14 | Y15 | Y16 | Y17 | Y18 | Y19 | Y20 | |
| 5 | Timing | ||||||||||||||||||||||
| 6 | Year | # | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | |
| 8 | The company | ||||||||||||||||||||||
| 9 | Discount rate | % | 12.0% | ||||||||||||||||||||
| 10 | Tax rate once profitable | % | 21.0% | ||||||||||||||||||||
| 11 | Diluted shares (millions) | m | 80.0 | ||||||||||||||||||||
| 12 | Cash and investments | $m | 320.0 | ||||||||||||||||||||
| 13 | Debt | $m | 40.0 | ||||||||||||||||||||
| 14 | Share price today | $ | 9.50 | ||||||||||||||||||||
| 15 | Unallocated corporate costs, Year 1 | $m | 35.0 | ||||||||||||||||||||
| 16 | Growth in corporate costs, % a year | % | 3.0% | ||||||||||||||||||||
| 18 | Stage probabilities of success | ||||||||||||||||||||||
| 19 | Phase 1 to Phase 2 | % | 60.0% | ||||||||||||||||||||
| 20 | Phase 2 to Phase 3 | % | 35.0% | ||||||||||||||||||||
| 21 | Phase 3 to filing | % | 60.0% | ||||||||||||||||||||
| 22 | Filing to approval | % | 90.0% |
Click any cell. The inspector names the line, the schedule it belongs to and what kind of cell it is; blue on cream is an input, black a calculation, green a value from another sheet. Scroll sideways to see every year.
Download
Biotech rNPV and sum-of-the-parts model: the workbook
Native Excel, formulas live, no macros, no external links. Inspect it above first; the file is the same model with the formulas in it.
A free account chooses one worked model to keep, and downloads every starter workbook. L3VLUP Pro ($25/month) opens the whole library. Signing in takes one email and no password.
What the base case says
- Programme A rNPV
- $712.6m
- Programme B rNPV
- $-32.3m
- Programme C rNPV
- $185.0m
- Equity value
- $810.7m
- Equity value per share
- 10.13
- Against the price today
- 6.7%
- Value that is probability
- $3,093.2m
Read from the workbook as served, every input at its default. Periods: Y1, Y2, Y3, Y4, Y5, Y6, Y7, Y8, Y9, Y10, Y11, Y12, Y13, Y14, Y15, Y16, Y17, Y18, Y19, Y20. The figures are invented and move with whatever you type in.
What this model is
A development-stage biotech with three programmes at three stages, each forecast from launch through loss of exclusivity: an addressable population, the share treated, a penetration ramp, a net price and an erosion at exclusivity loss.
Each programme’s cash flows are risk-adjusted by the probability of reaching the market from its current stage, the product of the stage probabilities that remain, with the current stage’s costs left unweighted because they are being spent regardless.
The sum of the parts: the three rNPVs, less the present value of corporate costs, plus net cash, to an equity value per share against the price today, with a discount-rate sensitivity in live formulas and a line that says how much of the value is probability.
Seats: Equity research and hedge funds, Investment banking, Private equity.
How the schedules connect
Every row in the workbook is tagged with the schedule it belongs to; the inspector above shows the tag when you click a row. These are the schedules this model is made of and where each one lives.
Clinical probability and stage gating
The chance of reaching the market, and what it weights
Assumptions, rows 19–22 · Programmes, rows 6–9 · Programme A, rows 6–8 · Programme B, rows 6–8 · Programme C, rows 6–8
Launch curve
Patients times price, ramped to peak, eroded at exclusivity loss
Programmes, rows 12–18 · Programme A, rows 11–16 · Programme B, rows 11–16 · Programme C, rows 11–16
Programme rNPV
Revenue to risk-adjusted cash flow to a value
Assumptions, row 6 · Programmes, rows 21–24 · Programme A, rows 19–32 · Programme B, rows 19–32 · Programme C, rows 19–32
Sum of the parts
Programmes, corporate costs, net cash, per share
Assumptions, rows 9–16 · SOTP, rows 6–9 · SOTP, rows 12–22 · SOTP, rows 25–28
What you should be able to explain
- Why the probability of reaching the market is a product of stage probabilities, and why the current stage’s costs are not risk-adjusted.
- How a launch curve is built from an addressable population, a treated share, a penetration ramp and a net price, and why gross-to-net matters.
- What loss of exclusivity does to a programme’s value and why the year it falls is one of the most sensitive inputs.
- Why corporate costs and net cash sit outside the programmes, and how much of a biotech’s value is probability rather than cash flow.
- Why one discount rate is used here and the stage risk is carried by the probability, not the rate.
What a reviewer looks for
- Risk-adjusting revenue but not the later-stage costs, so a failed Phase 2 pays for a Phase 3.
- A revenue line grown at a rate rather than built from patients, penetration and price.
- Gross price used as net, or exclusivity loss ignored, which can double a programme’s value.
- Stage risk carried twice, in a higher discount rate and in the probability.
- The sum of the parts quoted without the corporate costs that every programme has to pay for.
Conventions this workbook uses
Stated on the cover sheet too. A model is only as trustworthy as the decisions it tells you it made.
- Revenue is patients times price: addressable population, share diagnosed and treated, penetration ramping linearly to its peak over a set number of years, a gross price less the gross-to-net discount. After loss of exclusivity a retained share of revenue continues.
- The probability of reaching the market is the product of the stage probabilities still ahead of the programme. Cash flows inside the current stage carry a weight of one; everything after it carries the probability. Later-stage costs are therefore risk-adjusted too, which is the convention that keeps a failed Phase 2 from paying for a Phase 3.
- Tax is charged on programme profits only; losses shelter nothing in this model. A real company carries losses forward, which raises the value; the simplification is conservative and stated.
- Corporate costs are the unallocated cost of being a company and are not risk-adjusted: they are paid whether or not any programme succeeds.
- Discounting is mid-year at one company-wide rate. Practitioners sometimes discount earlier-stage programmes at higher rates; here the stage risk is carried by the probability, not the rate, so that the two are not double counted.
Build it yourself
The starter workbook
The launch curve has been cleared on every programme sheet: on the market, the penetration ramp, patients, net price, the exclusivity factor and net revenue. Build it from the Programmes sheet so that the cash flows, the rNPVs and the sum of the parts come back to life. The Checks sheet tells you when peak revenue reconciles to population, penetration and price.
Blanks: Launch curve. Free with any account. Compare with the worked model when you are done: download above.
The path around this model
Understand it, drill it, read the build, then apply it to a real company.
Read · Guide · 12 min
Biotech Valuation: rNPV, Launch Curves and the Sum of the Parts
Read · Guide · 12 min
How to Build a DCF Model: Build Order, Terminal Value and the Sanity Checks
Apply · Skill
Sum-of-the-Parts
Value the segments separately and show what a single multiple hides.
Apply · Skill
Model Audit
Find the errors in a financial model before someone senior does.
Vocabulary: rNPV (Risk-Adjusted Net Present Value), Probability of Technical and Regulatory Success (PTRS), Peak Sales, Loss of Exclusivity (LOE), Sum-of-the-Parts (SOTP), Gross-to-Net.
Questions about this model
Where do the stage probabilities come from?
Published industry base rates by phase, adjusted for the therapeutic area, the mechanism and the quality of the data so far. They are the most argued-over inputs in the model, which is why they sit together on the Assumptions sheet and each programme’s probability of reaching the market is shown as their product.
Why are the current stage’s costs not risk-adjusted?
Because the company is spending that money now, whatever the trial shows. Everything after the current stage, the next trial, the launch, the revenue, happens only if the current stage succeeds, so it is weighted by the probability. That is what makes an early programme cost more than it is worth until the data arrive.
How much of the value is probability?
The SOTP sheet says: the unrisked NPV of the programmes less their rNPV. For a pipeline with an early asset the gap is most of the value, which is why a single trial readout can move the share price more than any change in the discount rate.
Should each programme have its own discount rate?
Some practitioners use higher rates for earlier programmes. This model uses one rate and lets the probability carry the stage risk, because using both counts the same risk twice. The discount-rate sensitivity on the SOTP sheet shows what a different rate would do.
What does it cost?
Nothing to inspect: the whole workbook is on this page. A free account chooses one worked model to keep, and downloads every starter workbook. L3VLUP Pro ($25/month) downloads the whole library and opens every model’s formulas in the inspector.