# Privacy-Robust Incrementality Measurement for Advertising Systems under Signal Loss

arXiv: https://arxiv.org/abs/2606.03878 · HTML: https://mlsanity.com/paper/2606.03878/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 50 references that are in the map.
- First posted on arXiv 2 Jun 2026.

## Builds on (most cited first, 24 of 24)

- [Doubly robust off-policy evaluation with shrinkage](https://mlsanity.com/paper/1907.09623/) (2019)
- [Confounding-Robust Policy Improvement](https://mlsanity.com/paper/1805.08593/) (2018)
- [Attribution Modeling Increases Efficiency of Bidding in Display Advertising](https://mlsanity.com/paper/1707.06409/) (2017)
- [Causal Inference with Corrupted Data: Measurement Error, Missing Values, Discretization, and Differential Privacy](https://mlsanity.com/paper/2107.02780/) (2021)
- [Differentially Private Estimation of Heterogeneous Causal Effects](https://mlsanity.com/paper/2202.11043/) (2022)
- [Nonparametric Bounds and Sensitivity Analysis of Treatment Effects](https://mlsanity.com/paper/1503.01598/) (2015)
- [A Large Scale Benchmark for Individual Treatment Effect Prediction and Uplift Modeling](https://mlsanity.com/paper/2111.10106/) (2021)
- [CausalMTA: Eliminating the User Confounding Bias for Causal Multi-touch Attribution](https://mlsanity.com/paper/2201.00689/) (2021)
- [Challenges and approaches to privacy preserving post-click conversion prediction](https://mlsanity.com/paper/2201.12666/) (2022)
- [Locally Private Causal Inference for Randomized Experiments](https://mlsanity.com/paper/2301.01616/) (2023)
- [Causally Driven Incremental Multi Touch Attribution Using a Recurrent Neural Network](https://mlsanity.com/paper/1902.00215/) (2019)
- [Treatment Targeting by AUUC Maximization with Generalization Guarantees](https://mlsanity.com/paper/2012.09897/) (2020)
- [Nonparametric bounds for causal effects in imperfect randomized experiments](https://mlsanity.com/paper/2010.05220/) (2020)
- [Robust Causal Inference for Incremental Return on Ad Spend with Randomized Paired Geo Experiments](https://mlsanity.com/paper/1908.02922/) (2019)
- [On the Differential Privacy and Interactivity of Privacy Sandbox Reports](https://mlsanity.com/paper/2412.16916/) (2024)
- [Counterfactual-based Incrementality Measurement in a Digital Ad-Buying Platform](https://mlsanity.com/paper/1705.00634/) (2017)
- [Designing Experiments to Measure Incrementality on Facebook](https://mlsanity.com/paper/1806.02588/) (2018)
- [Latent Stratification for Incrementality Experiments](https://mlsanity.com/paper/1911.08438/) (2019)
- [Summary Reports Optimization in the Privacy Sandbox Attribution Reporting API](https://mlsanity.com/paper/2311.13586/) (2023)
- [Differentially Private Ad Conversion Measurement](https://mlsanity.com/paper/2403.15224/) (2024)
- [Cookie Monster: Efficient On-device Budgeting for Differentially-Private Ad-Measurement Systems](https://mlsanity.com/paper/2405.16719/) (2024)
- [Click Without Compromise: Online Advertising Measurement via Per User Differential Privacy](https://mlsanity.com/paper/2406.02463/) (2024)
- [Support-aware offline policy selection for advertising marketplaces](https://mlsanity.com/paper/2605.21736/) (2026)
- [Choosing Online Experiment Designs under Interference in Ads, Recommendations, and Member-Experience Systems](https://mlsanity.com/paper/2605.25290/) (2026)

## Built on by (most cited first, 1 of 1)

- [What a Data Release Must Preserve for Off-Policy Evaluation](https://mlsanity.com/paper/2610.15283/) (2026)

## Abstract (arXiv)

Advertising platforms use randomized lift tests to measure incrementality, but privacy-preserving reporting systems degrade the observed signal through match-rate loss, linkability loss, attribution-window loss, aggregation-threshold suppression, randomized reporting noise, and segment-heterogeneous signal loss. This paper formulates privacy-constrained advertising measurement as a robust causal decision problem under the mentioned signal losses. Given a randomized experiment and an ambiguity set for privacy-induced degradation, the framework projects the observation-compatible fiber of clean/unfiltered experimental worlds onto the incrementality functional and returns certified, rejected, and unresolved decisions. The main result gives a sharp decision frontier. Reports outside the frontier support uniformly valid certification or rejection, whereas reports inside it contain too little information for any method to uniformly distinguish above-threshold incrementality from non-incrementality. Supporting results give finite-sample certification, sample-complexity guarantees, a minimax lower bound showing that signal loss reduces effective information, and a reporting-granularity tradeoff. On 2.0M Criteo Uplift rows and the 64K-row Hillstrom email experiment, clean conversion lift is positive in both datasets, with lifts 0.00112 and 0.00495, respectively. Population certification survives mild degradation in Criteo and severe degradation in Hillstrom, while all considered finite-sample stress settings in both datasets remain unresolved after simultaneous uncertainty and reporting noise are included. Overall, the research contributes a decision-theoretic layer for privacy-aware incrementality measurement whose output is the strongest causal-claim justified by degraded ads signals.