LiDDA: Data Driven Attribution at LinkedIn

TL;DR

This paper presents LiDDA, a transformer-based data-driven attribution method at LinkedIn that effectively handles various data granularities and external factors, significantly improving marketing attribution accuracy.

Contribution

It introduces a unified transformer-based approach for attribution that integrates member-level, aggregate data, and macro factors, with large-scale implementation insights.

Findings

Significant impact on attribution accuracy at LinkedIn

Effective handling of diverse data granularities

Broadly applicable insights for marketing and ad tech

Abstract

Data Driven Attribution, which assigns conversion credits to marketing interactions based on causal patterns learned from data, is the foundation of modern marketing intelligence and vital to any marketing businesses and advertising platform. In this paper, we introduce a unified transformer-based attribution approach that can handle member-level data, aggregate-level data, and integration of external macro factors. We detail the large scale implementation of the approach at LinkedIn, showcasing significant impact. We also share learning and insights that are broadly applicable to the marketing and ad tech fields.