Whenever you're ready, walk me through how you'd approach this.
How would you determine if a marketing campaign actually caused an increase in sales versus just correlation?
Explain the identification strategy you would choose, the data and assumptions required, and how you would quantify uncertainty. Address what you would do if random assignment were unavailable, including possible confounding, seasonality, spillovers, pre-trends, and campaign timing. Describe the decision rule for claiming a causal lift and how you would assess practical significance.