Katkat
Back

How would you decide to cancel a TV show?

Capital OneData Scientist · Technical Screen
EasyData ScientistTechnical ScreenAnalytics & Experimentation

Last updated

This question tests the ability to define a structured framework for business decision-making by balancing quantitative metrics, such as viewership trends and production costs, with qualitative strategic considerations.

Scenario

You are a Data Scientist at a major streaming platform. The executives are concerned about the rising production costs of a long-running original series and are debating whether to cancel the show after the upcoming season or renew it for another.

Task

Design a data-driven framework to determine whether the show should be canceled or renewed. Your response should outline:

  1. The core metrics and KPIs you would track to measure the show’s performance.
  2. How you would account for both direct revenue (e.g., subscription growth) and indirect value (e.g., brand affinity).
  3. The approach you would take to estimate the 'churn risk' associated with canceling the show.
  4. Potential experiments or analyses you could run to validate your recommendation.

Hints

Hint 1

Consider the difference between 'acquisition' (the show bringing in new users) and 'retention' (the show preventing existing users from canceling).

Hint 2

Think about the cost side of the equation: how do you calculate the marginal profit per user generated by this specific show relative to its production budget?

Hint 3

Evaluate the 'halo effect' of the show, such as its impact on social media buzz, awards credibility, or the platform's overall brand perception.

Follow-up questions

  • How would you adjust your decision framework if the show is a 'niche' series with a very high engagement rate but a small total audience?
  • If the show's production costs are increasing by 20% annually, how would you model the break-even point for future seasons?
  • What qualitative data sources (e.g., social sentiment, critic reviews) would you integrate to augment your quantitative findings?