Pharma cases sit on the manufacturer side of healthcare: molecules, patents, payers, and sales forces. If the client is a hospital, you are on the wrong page.
Last full pass: 24 August 2026.
A pharmaceutical case interview in 2026 is a timing problem disguised as a marketing problem. The asset is a patent-protected cash engine whose net price is a rebate-and-coverage negotiation, not a list sticker, and whose volume is a treated-patient funnel (diagnosed × treated × share × days on therapy), not “units sold at Walmart.” McKinsey, BCG, Bain, L.E.K., ZS, and IQVIA all staff this work. Your job in the first 30 seconds is to name the decision: launch / don’t launch, price / access tradeoff, indication sequence, or what to do 18 months before loss of exclusivity (LOE). Provider operations belong on the healthcare case interview page. Devices, diagnostics, and mixed pipelines belong on life sciences.
What the interviewer wants you to decide
- Launch excellence: which indication, which country wave, how large a sales force, what the year-1 forecast should be.
- Pricing and market access: list vs net, rebate to PBMs / national payers, step-edits, and whether a lower net price buys a preferred tier.
- Pipeline / indication sequencing: a second indication can extend the franchise or starve the first launch of supply and attention.
- Patent cliff / LOE: authorized generic, next-gen molecule, geographic rest-of-world push, or harvest cash.
Clarify: brand vs generic vs biosimilar; US vs EU5 vs rest of world (US net price usually dominates value); specialty vs primary care (sales-force math changes); and whether “price” in the exhibit is WAC, invoice, or net after rebates, copay cards, and 340B.
Exhibits you should expect
| Exhibit | Trap |
|---|---|
| Treated-patient funnel | Prevalence is not treated patients. Diagnosis rate and persistency leak volume. |
| Gross-to-net bridge | A 9% list increase can be a 2% net decline if rebates and copay offset catch up. |
| Analog launch curves | Peak year is often year 4–6, not year 1. Front-loading a forecast is a classic miss. |
| Payer mix / formulary status | Unrestricted vs step-edit vs prior auth changes the realized share, not the “fair share” on a slide. |
| Remaining exclusivity | Value is cash flows until LOE, discounted. A beautiful peak sales number after patent expiry is decoration. |
| COGS vs SG&A vs R&D | COGS is often small; the P&L is commercial and science. Cutting “cost” like a CPG plant misses the point. |
Units and regulations that trip people
Patients vs scripts vs mg. Oncology is patients × duration × vials. Primary-care diabetes is scripts × days’ supply. Do not mix.
List vs net (gross-to-net). US list (WAC) is a fiction for profitability. IRA negotiation, Medicaid best price, 340B, and PBM rebates sit between WAC and cash. If you grow volume with a copay card, you may be buying demand with your own margin.
Patent cliff. After LOE, volume often holds better than price. A 70% volume retention at 20% of net price is a 86% revenue collapse, not a “share defense.”
FDA / EMA vs payer. Approval is not access. A drug can be indicated and still sit behind a step-edit for 18 months.
Practice a manufacturer-side pharma case
Work a scored case where net price, remaining exclusivity, and the patient funnel have to show up in the recommendation.
Worked mini-case: do not cut net price into a cliff
Prompt. Oncovance is a specialty oncology brand with 80,000 treated patients and $9,000 net per patient per year → $720m net sales (case inputs). LOE is in 18 months. The CFO wants a 15% net price cut now to “lock in share,” arguing volume would rise 10%. An authorized-generic plan would keep 30% of the franchise net in the first generic year at a $12m one-off cost. A second indication would take $90m and three years (after LOE). What do you do before the cliff?
Structure. (1) Cash until LOE, (2) price-cut arithmetic, (3) LOE-year scenarios, (4) what cannot earn its keep after the patent dies.
Math. Price cut: new net = $7,650; patients = 88,000; sales = $673m. That is −$47m vs $720m before you train every payer that this brand discounts. You also lower the Medicaid best-price reference and make the analog for the authorized generic worse.
LOE-year sketch (interview-style, stated assumptions): 70% of patients remain, net price falls to 20% of brand net → 80,000 × 0.70 × $1,800 = $101m. A 15% pre-cliff cut does not change that gravity.
Authorized generic: assume $720m × 30% = $216m in year 1 post-LOE at $12m cost. Ugly compared with the brand year, but it is cash you can actually collect, unlike a second indication that pays off after the cliff and still needs a trial win.
Recommendation. Do not cut net price. Harvest the brand: protect specialty access, freeze the sales-force expansion, and stand up the authorized-generic supply path. Kill or postpone the second indication unless it has a new patent or a label that surviving specialists will still use. Risk: a competitor launches a next-gen before LOE; that is a pipeline problem, not a 15% coupon. Next exhibit: gross-to-net bridge and remaining months of exclusivity by country.
What a generic profitability tree misses here
- Time is the product. A 20% margin improvement in year 5 is worthless if year 5 is generic.
- Price is net, after a rebate stack, not list × volume.
- Volume is a clinical funnel, not awareness. Diagnosis, eligibility, and persistency dominate advertising.
- COGS is the wrong cost conversation. Medical affairs, payer contracting, and pharmacovigilance are the real operating system.
- Regulation sets the feasible set. You cannot “just raise US price” into IRA / best-price physics.
If you treat Oncovance like a snack brand defending shelf space, you will recommend a discount into a patent expiry and call it share.
See where you stand on a pharma case
Practice LOE math and a net-price so-what before a ZS or L.E.K. round.
Related guides
- Healthcare case interview (providers, DRGs, beds)
- Life sciences consulting case interview
- Top pharma consulting firms
CoachNed is independent and unaffiliated with McKinsey, BCG, Bain, or other firms named for interview-style practice. Cases on CoachNed are AI-simulated.