Due to more stringent privacy regulations, fractured cross-platform data and disjointed advertising channels have reduced the traceability of digital advertising results, and now people are only focused on changes in scale. Existing research has not yet combined propensity score correction, counterfactual incremental estimation, Shapley decomposition and marginal iROAS constraints in an auditable attribution pipeline. A bias-corrected multi-touchpoint attribution framework is proposed in this paper to remove the natural conversion baseline and redistribute channel credit. As shown in the illustrative channel-level application, observed attribution can differ from corrected incremental contribution by as much as 11 percentage points; for example, retargeting marginal iROAS dropped to 0.72, and 17 per cent of the budget was shifted among channels. Based on the above results, the attribution share should not be directly correlated with true increment; thus, budget allocation should shift away from high-relevance but low-incremental harvesting channels and towards scalable channels that offer better marginal incremental returns.

Yiyu Yang. Identification of True Incremental Effects and Budget Reallocation Mechanisms in Digital Advertising under Bias-Corrected Multi-Touch Attribution. Socio-Economic Statistics Research (2026), Vol. 7, Issue 1: 19-26. https://doi.org/10.38007/SESR.2026.070203.

[1] Hosahally S. Measuring digital advertising in a post-cookie era: A study of marketing-mix models, attribution and incrementality. Journal of Digital & Social Media Marketing , 2025, 12(4): 1-13.

[2] Gordon BR, Moakler R, Zettelmeyer F. Predicted Incrementality by Experimentation (PIE) for Ad Measurement. arXiv preprint arXiv:2304.06828 , 2023/2026.

[3] Runge J, Skokan I, Zhou G, Pauwels K. Packaging Up Media Mix Modeling: An Introduction to Robyn's Open-Source Approach. arXiv preprint arXiv:2403.14674 , 2025.

[4] Gangopadhyay B, Wang Z, Chiappa AS, Takamatsu S. Adaptive Budget Optimization for Multichannel Advertising Using Combinatorial Bandits. arXiv preprint arXiv:2502.02920 , 2025.

[5] Waisman C, Gordon B R. Multi-cell Experiments for Marginal Treatment Effect Estimation of Digital Ads. arXiv preprint arXiv:2302.13857 , 2025.

[6] Wu, Y. (2026). Research on Interpretable Confidence Rule Base Modeling Method Integrating Data-Driven and Constrained K-Means Optimization. Procedia Computer Science, 279, 612-619.

[7] Sun, J. (2026). Automated Feature Engineering and Screening System for Large-Scale Factor Libraries. Procedia Computer Science, 281, 1282-1290.

[8] Zhang, Z. (2026). Research on Performance Monitoring and Data-driven Optimization Mechanisms for AI Systems Oriented Toward Decision Support. Advances in Computer and Communication, 7(2).

[9] Hou, Y. (2026). Collaborative Regulation for Stable Data Center Operation under Energy Efficiency Constraints. Procedia Computer Science, 281, 612-620.

[10] Wu, Y. (2026). Research on Interpretable Confidence Rule Base Modeling Method Integrating Data-Driven and Constrained K-Means Optimization. Procedia Computer Science, 279, 612-619.

[11] Zhang, Z. (2026). Research on Performance Optimization Methods for Resource-Aware Model Services in AI Systems. Procedia Computer Science, 281, 1310-1317.

[12] Liu, Z. (2026). Research on Measuring User Behavior Response Differences Supported by Propensity Scoring Method. European Journal of AI, Computing & Informatics, 2(2), 47-53.

[13] Gao, Y. (2026). Governance Structures and Checks and Balances in Private Equity Funds: Practice, Problems, and Improvement Paths. Financial Economics Insights, 3(2), 82-91.

[14] Wu, L. (2026). Construction and Evolutionary Analysis of a Game Model for Supply Chain Finance Funding Based on Blockchain Technology. Procedia Computer Science, 282, 2004-2012.