Evaluate Account-Partner Performance with Observational Data Analysis

##### Scenario DoorDash uses account partners to acquire new merchants and needs to evaluate the program’s effectiveness ##### Question a) What success metrics would you track to judge account-partner performance for merchant onboarding? b) Without an A/B test, how would you use observational data (e.g., difference-in-differences) to estimate lift? c) If account partners recruit mainly seafood restaurants, how would you create a comparable control group (e.g., propensity-score matching, synthetic control) and what covariates would you include? d) How would you model the relationship between partner-acquisition spend and incremental revenue, and why might a logistic form be more appropriate than a linear one? ##### Hints Think causal inference: parallel trends, matching, key attributes; include geo, cuisine, traffic. Explain diminishing returns in spend–response curve.