arXiv 22 May 2026 · Econometrics
arXiv:2605.23706 · PDF · DOI · OpenAlex · Extracted main text
Online advertising platforms host hundreds of thousands of A/B tests, but the platform's delivery algorithm routes each creative to the audience it predicts will engage. Every two-arm test therefore conflates the creative's effect with the algorithm's targeting response, and adjusting for the realized audience is biased because audience is a post-treatment mediator. We propose a three-arm design that adds an arm exposing the algorithm to the treatment metadata while holding the user-facing creative identical to control, point-identifying the natural indirect (algorithmic) and direct (creative) effects without sequential ignorability. In a live Meta campaign with a women-targeted text fragment, the algorithmic channel raises female impression share by +2.07 ppt while the creative channel moves it by -0.68 ppt; roughly three-quarters of the absolute reallocation is algorithmic, and a conventional two-arm test understates the algorithmic channel by a factor of two. The design isolates the contribution of platform's algorithm to the outcome which is separable from creative content.
appendix boundary found by appendix_command · 70% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Judea Pearl (2001) Direct and indirect effects | 1.000 | 7 | 3 | 100% |
| 2 | Alan S. Gerber and Donald P. Green (2012) Field Experiments: Design, Analysis, and Interpretation | 1.000 | 6 | 3 | 100% |
| 3 | Kosuke Imai, Dustin Tingley, and Teppei Yamamoto (2013) Experimental designs for identifying causal mechanisms | 1.000 | 6 | 3 | 100% |
| 4 | Gordon Burtch, Robert Moakler, Brett R. Gordon, Poppy Zhang, and Sha… (2025) Characterizing and minimizing divergent delivery in Meta advertising experiments | 0.956 | 8 | 4 | 88% |
| 5 | Avidit Acharya, Matthew Blackwell, and Maya Sen (2016) Explaining causal findings without bias: Detecting and assessing direct effects | 0.928 | 4 | 3 | 100% |
| 6 | James M. Robins and Sander Greenland (1992) Identifiability and exchangeability for direct and indirect effects | 0.928 | 4 | 3 | 100% |
| 7 | Tyler J. VanderWeele (2015) Explanation in Causal Inference: Methods for Mediation and Interaction | 0.811 | 4 | 2 | 100% |
| 8 | Muhammad Ali, Piotr Sapieżyński, Miranda Bogen, Aleksandra Korolova,… (2019) Discrimination through optimization: How Facebook's ad delivery can lead to biased outcomes | 0.737 | 3 | 2 | 100% |
| 9 | Reuben M. Baron and David A. Kenny (1986) The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerati… | 0.737 | 3 | 2 | 100% |
| 10 | Kosuke Imai, Luke Keele, and Dustin Tingley (2010) A general approach to causal mediation analysis | 0.737 | 3 | 2 | 100% |
Showing the top 10 of 36 scored citations.