Amazon Ads Multi-Touch Attribution

TL;DR

Amazon's Multi-Touch Attribution combines randomized controlled trials and machine learning to accurately measure each ad touchpoint's contribution to conversions, providing advertisers with a comprehensive performance view.

Contribution

This paper introduces a novel methodology that integrates RCTs, ML models, and shopping signals for improved ad attribution accuracy.

Findings

ML models offer scalable, precise predictions but may be biased.

RCTs provide unbiased effects but are noisy.

Combined approach enhances attribution reliability.

Abstract

Amazon's new Multi-Touch Attribution (MTA) solution allows advertisers to measure how each touchpoint across the marketing funnel contributes to a conversion. This gives advertisers a more comprehensive view of their Amazon Ads performance across objectives when multiple ads influence shopping decisions. Amazon MTA uses a combination of randomized controlled trials (RCTs) and machine learning (ML) models to allocate credit for Amazon conversions across Amazon Ads touchpoints in proportion to their value, i.e., their likely contribution to shopping decisions. ML models trained purely on observational data are easy to scale and can yield precise predictions, but the models might produce biased estimates of ad effects. RCTs yield unbiased ad effects but can be noisy. Our MTA methodology combines experiments, ML models, and Amazon's shopping signals in a thoughtful manner to inform attribution credit allocation.