LiDDA: Data Driven Attribution at LinkedIn

📰 ArXiv cs.AI

Learn how LinkedIn's LiDDA uses a unified transformer-based approach for data-driven attribution in marketing intelligence

advanced Published 28 May 2026

Action Steps

Build a data-driven attribution model using a transformer-based approach
Integrate member-level and aggregate-level data into the model
Incorporate external macro factors to improve causal pattern learning
Configure the model to handle large-scale data
Test the model's performance using metrics such as conversion credits and marketing ROI
Apply the model to real-world marketing data to inform attribution decisions

Who Needs to Know This

Data scientists and marketers on a team can benefit from understanding LiDDA to improve marketing intelligence and attribution modeling

Key Insight

💡 A unified transformer-based approach can effectively handle multiple data types and external factors for improved marketing attribution

Key Takeaways

Learn how LinkedIn's LiDDA uses a unified transformer-based approach for data-driven attribution in marketing intelligence

Full Article

Title: LiDDA: Data Driven Attribution at LinkedIn

Abstract:
arXiv:2505.09861v3 Announce Type: replace-cross 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 business 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 scal

Read full paper → ← Back to Reads