## Case: Should you cancel, keep, or sell a TV series? You are the **CEO of a streaming company**. Your platform currently produces and distributes a scripted TV series called **"Analyst."** A second project, **"Shark Bank,"** is on the table as an alternative use of the same production budget and slot. Throughout the case the interviewer will move between qualitative framework questions and back-of-the-envelope calculations, so be ready to switch between structured reasoning and numbers. Your objective as CEO is to **maximize long-term company value**, not just the current quarter's profit. ### Constraints & Assumptions State these explicitly when you answer; flag any you would confirm with the interviewer. - **Objective:** maximize long-term, risk-adjusted enterprise value (subscriber base + IP + brand), not single-quarter profit. - **Profitability** means **incremental contribution attributable to the show** — including downstream subscription effects (acquisition + retention), not just the show's direct ad/licensing revenue. - You do **not** have a detailed P&L; cost reasoning is "common sense" unless the interviewer hands you figures. - When financials are needed you may use **NPV** with a reasonable discount rate plus scenario analysis. The interviewer may instead give you a short, fixed project horizon and ask for a simple total-profit comparison — use whichever framing the exhibit supports. - For the comparative calculation (Part 2) and the sell decision (Part 5), the interviewer supplies numbers verbally or via an exhibit. **Do not assume probabilities or fill in missing inputs — ask for them.** ### Clarifying Questions to Ask Ask these up front to scope the whole case before diving into any single part. - What is the **decision horizon**? (e.g., is "Analyst" a finite-life project — 2 more years — or open-ended renewal?) - What is our **business model / revenue mix**? Subscription-only (SVOD), ad-supported (AVOD), or hybrid? This changes which revenue drivers matter. - How do we currently **attribute subscribers** to a specific show (acquisition and retention)? Do we have viewer-to-subscriber linkage? - What is the **alternative** use of the budget/slot? (Here: "Shark Bank" — what is its profile and risk?) - For any expected-value comparison, **what are the success/failure probabilities and the payoffs** for each project? (Never assume 50/50.) - Are there **constraints** I should weight — capital/liquidity needs, production capacity, existing talent contracts, brand considerations? ### Part 1 — Cancel vs. renew: what factors would you evaluate? You are considering **canceling** "Analyst." (1) What factors would you weigh in the cancel-vs-renew decision? (2) What **market/competitive** considerations (genre trends, competitor moves) might change the decision? ```hint Frame it as incremental value Don't list revenue and cost in isolation — frame the decision as the *incremental* enterprise value of renewing vs. canceling. The counterfactual matters: if you cancel, where do those viewers go (churn, watch other content on your platform, or switch to a competitor)? ``` ```hint The biggest line item is often invisible For a subscription business the largest value of a show is frequently NOT its direct ad/licensing revenue but its effect on **subscriber acquisition and retention**. Make sure your factor list separates direct revenue from subscription-driven value. ``` ```hint Don't forget the second question Market context is a separate, explicitly-asked dimension: genre trajectory (growing/declining), competitor slate launching substitutes, defensibility/uniqueness of the show, and the ad/licensing demand environment. ``` #### What This Part Should Cover - A clear **incremental / counterfactual framing** (value of the show = what changes *because* it exists). - Separation of **direct revenue**, **subscription-driven value** (acquisition + retention + engagement), and **costs**. - Concrete **metrics to pull** (viewership, completion, cohort conversion, churn hazard for exposed vs. unexposed) — not just vague "look at the data." - An explicit treatment of the **market/competitive** dimension (genre trend, competitor moves, defensibility). ### Part 2 — Comparative expected return: "Analyst" vs. "Shark Bank" The interviewer gives you an exhibit with the expected payoffs for each project. Compute the **expected return of each** and recommend which to greenlight. ```hint Ask before you compute The exhibit does NOT give you the probabilities. Explicitly ask the interviewer for **"Shark Bank's" chance of success vs. failure** (and "Analyst's"). Do not silently assume 50/50 — that assumption is the trap in this part. ``` ```hint The formula For each project compute $E[\text{return}] = \sum_i p_i \cdot v_i$ over its outcome states (e.g., $p_{\text{success}}\cdot v_{\text{success}} + p_{\text{fail}}\cdot v_{\text{fail}}$), net of cost. Compare the two expected values, then sanity-check against risk/variance, not just the point estimate. ``` #### What This Part Should Cover - **Eliciting the missing probabilities** before calculating (the explicit ask). - Correct **expected-value arithmetic** for both projects. - A recommendation that compares the EVs **and** acknowledges risk/variance and any shared-resource constraint (same budget/slot). ### Part 3 — How would you increase "Analyst's" profitability? Assume you keep the show for now. (1) What are the major **revenue drivers**? (2) What are the major **cost drivers** (common-sense breakdown — you don't have the P&L)? (3) Propose **levers** to lift profit (revenue up and/or cost down). ```hint Decompose before you brainstorm Write a profit decomposition first, then attach levers to each term. Roughly: $\Delta\text{Profit} \approx \Delta\text{Subs}\times\text{LTV} + \Delta\text{AdRev} + \Delta\text{Licensing} - \Delta\text{Production} - \Delta\text{Marketing}$. Each term suggests its own lever. ``` ```hint Prepare the cost side too A common stumble here is being fluent on revenue but vague on cost. Be ready to name a sensible cost structure from first principles: above-the-line talent (cast, writers, music rights), below-the-line production (crew, sets, VFX, post), marketing/PR, and distribution/streaming ops. Don't claim you "can't guess" — reason from common sense. ``` #### What This Part Should Cover - A **decomposition** of revenue (direct + subscription-driven) and cost (above-the-line, below-the-line, marketing, distribution). - **Levers** tied to each term, with a rough sense of which have the largest, most testable upside. - An idea of **how to prioritize** levers (scenario/sensitivity analysis, experiments where possible) rather than an unranked list. ### Part 4 — Should you sell the "Analyst" project / IP? You are now considering **selling "Analyst"** to another company. (1) What factors determine whether selling beats keeping? (2) What data/analyses would you request to make the recommendation? ```hint Three quantities to compare Reduce it to: sale **price today** vs. **NPV of keeping** (including the subscription impact) vs. **strategic/option value** not captured in near-term cash (franchise, spin-offs, brand). Sell when price exceeds NPV(keep) *and* you don't need the strategic option. ``` #### What This Part Should Cover - The **three-quantity comparison** (price vs. NPV-of-keep vs. strategic/option value). - The **data you'd request** to estimate NPV-of-keep credibly (P&L + contracts, viewer-to-subscriber attribution, churn model, demand forecast, portfolio backfill). - Recognition that the answer is **conditional** on company circumstances (liquidity needs, risk tolerance) — not a single fixed yes/no. ### Part 5 — The concrete sell decision The interviewer now gives you specific numbers verbally: - "Analyst" has a **remaining life of 2 years**. - "Shark Bank" (the alternative project) would yield a **total profit of \$22M over 2 years**. - Selling "Analyst" would **lose 1.5M subscribers**, each worth **\$32** (over the relevant period). A competitor offers exactly your computed gap to buy "Analyst." **Do you sell?** Then: would your answer change if your company were **in financial distress / cash-constrained**? ```hint Build the comparison number Quantify the value of *keeping* "Analyst" relative to the best alternative. First state explicitly whether you're treating the two projects as competing for the same slot (mutually exclusive) or independent — that assumption determines the structure of your breakeven. Then show the arithmetic transparently and flag what you're leaving out. ``` ```hint What you can quantify isn't the whole picture Once you have a breakeven number, think carefully about what the quantified figure captures vs. what it omits. Consider whether there are sources of value — or sources of risk — that your arithmetic hasn't accounted for, and how their presence should affect your decision at the exact breakeven point. ``` #### What This Part Should Cover - An explicit statement of the **mutually-exclusive-slot assumption** (or a question to the interviewer confirming it) before building the breakeven. - Transparent **arithmetic** for the breakeven sale price, with stated inclusions/omissions. - A defensible **sell/keep call at exact breakeven** that invokes un-quantified upside (option value) — and recognizes there is no single "correct" number, only a well-reasoned one. - A coherent **change of recommendation under cash constraint** (time value / certainty of cash vs. uncertain future value). ### What a Strong Answer Covers These dimensions span all five parts; the interviewer is reading for them throughout, not just within any one part. - **Structured, MECE thinking** that consistently returns to *incremental, risk-adjusted long-term value* as the decision criterion. - **Data-scientist rigor on causality:** flagging that heavy viewers self-select (selection bias), that loyalty may cause viewing rather than the reverse (reverse causality), and proposing identification strategies (matched cohorts, natural experiments, uplift modeling) rather than naive correlations. - **Comfort switching between qualitative framework and quantitative back-of-envelope**, asking for missing inputs instead of assuming them. - **Executive judgment:** stating assumptions, giving a recommendation with a "what would change my mind" threshold, and adapting to constraints (distress, capacity, contracts). ### Follow-up Questions - In Part 2, suppose "Shark Bank" has much higher variance than "Analyst" even at a slightly higher expected value — how does that change your greenlight decision, and what would you tell the board? - How would you **causally** estimate the 1.5M-subscriber loss in Part 5 rather than taking it as given? What quasi-experiment or model would you use, and how would you bound the estimate? - If the competitor's offer were **30% above** your computed breakeven, what additional considerations (strategic, competitive, antitrust/brand) would still make you hesitate to sell?