Super Bowl National TV Ad: Incrementality Measurement Plan

Context: Your consumer app ran a single 30-second national Super Bowl TV spot. Design a plan to estimate incremental impact on installs and revenue without a clean randomized control.

Answer all parts with specificity:

(a) KPIs and Measurement Windows

• Define primary and guardrail KPIs.
• Specify exact measurement windows: pre-period, game day segmented by quarter, and post-period.

(b) Identification Strategies

• Propose at least two credible strategies (e.g., geo-level difference-in-differences using heterogeneous DMA viewership, synthetic control using similar apps/markets, matched markets, MMM with a spike regressor).

(c) DiD Specification and Assumptions

• Write the estimating equation, list assumptions (parallel trends, SUTVA, no differential shocks), and describe how you would test them.

(d) Confounders

• Plan for concurrent promotions, outages, app store featuring, and game-related seasonal spikes.

(e) Uncertainty

• Show how you will estimate and report uncertainty (e.g., robust/cluster-robust SEs, placebo tests, randomization inference).

(f) Validation

• Falsification checks (pre-trend, placebo Super Bowl dates) and heterogeneity cuts (new vs existing users, platforms).

(g) Power

• If the true lift is a 4% increase in daily installs on game day, list the data you need and compute an approximate minimum sample (markets and days) to detect it at 5% alpha and 80% power.