Towards Measuring Sell Side Outcomes in Buy Side Marketplace Experiments using In-Experiment Bipartite Graph
This addresses a crucial need for modern marketplaces to measure cross-side outcomes, though it is incremental by building on existing bipartite graph and mediation analysis methods.
The study tackled the problem of evaluating seller-side causal effects in buyer-side marketplace experiments by constructing a bipartite graph from in-experiment data, demonstrating it on Vinted's platform with over 80M users.
In this study, we evaluate causal inference estimators for online controlled bipartite graph experiments in a real marketplace setting. Our novel contribution is constructing a bipartite graph using in-experiment data, rather than relying on prior knowledge or historical data, the common approach in the literature published to date. We build the bipartite graph from various interactions between buyers and sellers in the marketplace, establishing a novel research direction at the intersection of bipartite experiments and mediation analysis. This approach is crucial for modern marketplaces aiming to evaluate seller-side causal effects in buyer-side experiments, or vice versa. We demonstrate our method using historical buyer-side experiments conducted at Vinted, the largest second-hand marketplace in Europe with over 80M users.