Examination of e-commerce data practices

Summary

This article explores the propagation of advertising and analytics identifiers—such as those from Google, Facebook, and AppLovin—across e-commerce ecosystems, emphasizing the role of third-party libraries like Elevar. Using a custom Firefox plugin, the author traces data flows on sites like crocs.com and thewoobles.com, finding that identifiers are commonly appended as cart attributes and sent to multiple partners, often on pre-checkout pages. A key insight is that blocking Elevar prevents identifier sharing, underscoring its role as a data conduit.

The article also cautions against false positives: for instance, an 'igId' attribute on trueclassictees.com appears Instagram-related but actually comes from Intelligems, a profit optimization tool. Additionally, seemingly random strings like AppLovin's connectEventKey are static pixel configurations, not user identifiers. The author concludes that privacy frameworks (e.g., ITP) have forced all companies into similar constraints, but understanding data flows requires deep technical context.

AppLovin’s approach is to use standard APIs and discard extraneous data, focusing on machine learning for sustainable growth. For ad ops decision-makers, the key takeaway is the need to audit third-party integrations to map identifier propagation, differentiate false positives from actual tracking, and ensure compliance with privacy norms while leveraging data for personalization.