About

Duncan Simester is a Professor at the MIT Sloan School of Management where he holds the NTU Chair in Management Science and serves as the Chair of the MIT Evening MBA Task Force. He is an expert on how economics and artificial intelligence can contribute to the understanding and practice of marketing and strategy. His work is widely published in the academic literature, often relying on industry participation and large-scale field experiments conducted with cooperating firms. Prior to joining MIT, Duncan was a professor at the University of Chicago’s Graduate School of Business. He is also a qualified lawyer and a member of the bar in New Zealand. Duncan holds a PhD from MIT and has received notable honors including the 2020 Weitz-Winer-O’Dell Award from the American Marketing Association and the 2022 INFORMS Society for Marketing Science Fellow Award, recognizing his significant contributions to marketing research, theory, methodology, and practice.

Research topics

• Computer Science
• Economics
• Business
• Management science
• Computer Security
• Political Science
• Mathematics
• Marketing
• Process management
• Monetary economics
• Data science
• Financial economics
• Finance
• Macroeconomics
• Management

Selected publications

• A Sample Size Calculation for Training and Certifying Targeting Policies

  Management Science · 2025-03-24 · 1 citations

  article1st authorCorresponding

  We propose an approach for determining the sample size required when using an experiment to train and certify a targeting policy. Calculating the rate at which the performance of a targeting model improves with additional training data is a complex problem. We address this challenge by assuming that customers are grouped into segments that capture relevant information about their responsiveness to the firm’s marketing actions. We consider two problem formulations. The first formulation identifies the sample size required to train a targeting policy and certify that its expected performance exceeds a predefined threshold. The second formulation identifies the sample size required to train a targeting policy and certify that it outperforms a baseline in an out-of-sample statistical test. We establish theoretical properties of these problems, based on which we propose computationally efficient algorithms for optimal sample size calculations. We illustrate our algorithms and analysis using data from a luxury fashion retailer. This paper was accepted by David Simchi-Levi, marketing. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.02947 .

  PublisherDOI

• Optimizing Scalable Targeted Marketing Policies with Constraints

  Marketing Science · 2025-03-20 · 2 citations

  article

  This paper introduces a novel optimization algorithm to address targeting problems with a large and complex set of constraints.

  PublisherDOI

• Improving Targeting Policies Using Transfer Learning

  SSRN Electronic Journal · 2025-01-01 · 1 citations

  preprintOpen access
  PublisherDOI

• Household Portfolios and Retirement Saving over the Life Cycle

  The Journal of Finance · 2025-08-12 · 5 citations

  articleOpen accessSenior author

  ABSTRACT Using account‐level data on millions of U.S. middle‐class investors over 2006 to 2018, we characterize the share of investable wealth that they hold in the stock market over their working lives. Relative to the 1990s, this share has both risen by 10% and become age‐dependent. The Pension Protection Act (PPA)—which allowed target date funds (TDFs) to be default options in retirement plans—played an important role: younger (older) workers starting at a firm after TDFs became the default option post‐PPA invested more (less) in stocks, in line with the TDF glidepath. In contrast, contribution rates changed little following the PPA.

  PublisherOA PDFDOI

• Targeted Marketing with Large Batches

  2024-01-01

  articleOpen accessSenior author
  PublisherDOI

• Optimizing Scalable Targeted Marketing Policies with Constraints

  arXiv (Cornell University) · 2023-12-02

  preprintOpen access

  Targeted marketing policies target different customers with different marketing actions. While most research has focused on training targeting policies without managerial constraints, in practice, many firms face managerial constraints when implementing these policies. For example, firms may face volume constraints on the maximum or minimum number of actions they can take, or on the minimum acceptable outcomes for different customer segments. They may also face similarity (fairness) constraints that require similar actions with different groups of customers. Traditional optimization methods face challenges when solving problems with either many customers or many constraints. We show how recent advances in linear programming can be adapted to the targeting of marketing actions. We provide a theoretical guarantee comparing how the proposed algorithm scales compared to state-of-the-art benchmarks (primal simplex, dual simplex and barrier methods). We also extend existing guarantees on optimality and computation speed, by adapting them to accommodate the characteristics of targeting problems. We implement the proposed algorithm using data from a field experiment with over 2 million customers, and six different marketing actions (including a no action ``Control''). We use this application to evaluate the computation speed and range of problems the algorithm can solve, comparing it to benchmark methods. The findings confirm that the algorithm makes it feasible to train large-scale targeting problems that include volume and similarity constraints.

  PublisherOA PDFDOI

• Optimizing Scalable Targeted Marketing Policies with Constraints

  SSRN Electronic Journal · 2023-01-01 · 3 citations

  articleOpen access
  PublisherDOI

• Canary Categories

  Journal of Marketing Research · 2023-11-29

  articleOpen accessSenior author

  Past customer spending in a category is generally a positive signal of future customer spending. Analyses of historical data at two retailers demonstrate that there exist “canary categories” for which the reverse is true. Purchases in these categories are a signal that customers are less likely to return to that retailer. The authors propose an explanation for the existence of canary categories and then develop a stylized model that illustrates four contributing factors: the probability that a customer finds their favorite brand, customers’ willingness to substitute brands, the cost and attractiveness of visiting other stores, and expectations about future brand availability. The analysis uses both field data and experiments to investigate these factors. The findings suggest that canary categories exist (at least in part) because store assortments are not completely adjusted to local preferences. An implication is that canary categories are endogenous to each retailer; the same category may be a canary category at one retailer and a destination category at a competing retailer.

  PublisherOA PDFDOI

• Household Portfolios and Retirement Saving Over the Life Cycle

  SSRN Electronic Journal · 2022-01-01 · 16 citations

  articleOpen accessSenior author
  PublisherDOI

• Price Frictions and the Success of New Products

  Marketing Science · 2022-07-26 · 8 citations

  articleSenior author

  Price frictions reduce the success of new products and impact retailers’ product assortments.

  PublisherDOI

Frequent coauthors

• Eric T. Anderson

  Northwestern University

  33 shared

• John R. Hauser

  Massachusetts Institute of Technology

  21 shared

• Florian Zettelmeyer

  16 shared

• Meghan R. Busse

  14 shared

• Birger Wernerfelt

  Massachusetts Institute of Technology

  13 shared

• Erik Brynjolfsson

  National Bureau of Economic Research

  13 shared

• Spyros I. Zoumpoulis

  Decision Sciences (United States)

  11 shared

• Olivier Toubia

  Columbia University

  8 shared

Awards & honors

• 2020 Weitz-Winer-O’Dell Award from the American Marketing As…
• 2022 INFORMS Society for Marketing Science (ISMS) Fellow Awa…