Evaluating online ad campaigns in a pipeline

TLDR

Display ads proliferate online, yet their effectiveness remains uncertain amid competing advertising. The authors present a rapid, accurate method that evaluates ad effectiveness without experiments, surveys, focus groups, or expert analysts. The method uses doubly robust estimation to counter selection bias and a nonparametric test on irrelevant outcomes for further defense. Simulations show the approach yields more robust estimates than regression or propensity scoring and supports fully automated, rapid processing from data retrieval to report generation.

Abstract

Display ads proliferate on the web, but are they effective? Or are they irrelevant in light of all the other advertising that people see? We describe a way to answer these questions, quickly and accurately, without randomized experiments, surveys, focus groups or expert data analysts. Doubly robust estimation protects against the selection bias that is inherent in observational data, and a nonparametric test that is based on irrelevant outcomes provides further defense. Simulations based on realistic scenarios show that the resulting estimates are more robust to selection bias than traditional alternatives, such as regression modeling or propensity scoring. Moreover, computations are fast enough that all processing, from data retrieval through estimation, testing, validation and report generation, proceeds in an automated pipeline, without anyone needing to see the raw data.