# The Value of Personalized Recommendations: Evidence from Netflix

arXiv: https://arxiv.org/abs/2511.07280 · HTML: https://mlsanity.com/paper/2511.07280/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 64 references that are in the map.
- First posted on arXiv 10 Nov 2025.

## Builds on (most cited first, 9 of 9)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches](https://mlsanity.com/paper/1907.06902/) (2019)
- [The Engagement-Diversity Connection: Evidence from a Field Experiment on Spotify](https://mlsanity.com/paper/2003.08203/) (2020)
- [Revealed Preference at Scale: Learning Personalized Preferences from Assortment Choices](https://mlsanity.com/paper/1509.05113/) (2015)
- [Studying Product Competition Using Representation Learning](https://mlsanity.com/paper/2005.10402/) (2020)
- [Learning the Covariance of Treatment Effects Across Many Weak Experiments](https://mlsanity.com/paper/2402.17637/) (2024)
- [Sliding Window Training -- Utilizing Historical Recommender Systems Data for Foundation Models](https://mlsanity.com/paper/2409.14517/) (2024)
- [Adventures in Demand Analysis Using AI](https://mlsanity.com/paper/2501.00382/) (2024)
- [Demand Estimation with Text and Image Data](https://mlsanity.com/paper/2503.20711/) (2025)

## Built on by (most cited first, 2 of 2)

- [Efficient Inference for Inverse Reinforcement Learning and Dynamic Discrete Choice Models](https://mlsanity.com/paper/2512.24407/) (2025)
- [Cost-Ordered Feasibility for Multi-Armed Bandits with Cost Subsidy](https://mlsanity.com/paper/2605.07171/) (2026)

## Abstract (arXiv)

Personalized recommendation systems shape much of user choice online, yet their targeted nature makes separating out the value of recommendation and the underlying goods challenging. We build a discrete choice model that embeds recommendation-induced utility, low-rank heterogeneity, and flexible state dependence and apply the model to viewership data at Netflix. We exploit idiosyncratic variation introduced by the recommendation algorithm to identify and separately value these components as well as to recover model-free diversion ratios that we can use to validate our structural model. We use the model to evaluate counterfactuals that quantify the incremental engagement generated by personalized recommendations. First, we show that replacing the current recommender system with a matrix factorization or popularity-based algorithm would lead to 4% and 12% reduction in engagement, respectively, and decreased consumption diversity. Second, most of the consumption increase from recommendations comes from effective targeting, not mechanical exposure, with the largest gains for mid-popularity goods (as opposed to broadly appealing or very niche goods).