This question evaluates a data scientist's skills in revenue modeling, funnel construction, causal inference for bias correction, experimentation scaling, and sensitivity analysis using organic engagement and purchase metrics.

You are evaluating the potential monthly revenue impact of launching a new Shopping tab in a social photo app. Only organic surfaces are in scope (organic posts and in-app search). Exclude ads and any paid placements.
You have access to the following data from current organic experiences:
Assume purchases are on-platform and can be linked to prior organic exposures.
Using the available data, estimate monthly potential revenue if the Shopping tab is launched to 100% of users. Provide:
(a) A clearly specified conversion funnel and algebraic formula for revenue.
(b) How you would obtain unbiased CTR/PDP rates from current organic entry points (given position/selection biases).
(c) How you would adjust for cannibalization of existing organic purchases.
(d) How you would scale from experiment traffic to the full population while accounting for heterogeneity.
(e) A sensitivity analysis on the key assumptions and parameters.
Login required