Choice Modeling

Utility Maximization vs. Regret Minimization

Mar 24, 2022

Choice Modeling: Utility Maximization & Regret Minimization
A deeper look into the major approaches to choice modeling and how consumers make choices.

STEVE COHENChoice modeling is one of the most insightful ways to uncover insights about consumer preferences for products, services, and experiences. There are various models out there, but not all of them work the same way — the diagram below outlines the four major approaches:

Choice Modeling How many items are chosen? How many units of each are chosen? Multinominal Logit Menu Model Quantity Discount Multiple Discrete Choice & Volume

The first cell in the upper left is the choice of One unit of one product/brand: This is your typical Choice-Based Conjoint Analysis (CBCA). CBCA has been a powerful tool for researchers, but there are other statistical models that can be more effective in certain situations.

Utility Maximization

The Multinomial Logit (MNL) method assumes that consumers are utility maximizers (i.e., they’re always looking to get the most enjoyment from their purchase).

The way the math of MNL works is: one choice at a time, the weight that the consumer places on each attribute is summed up and then the weight times price is subtracted. Then, MNL calculates the share of each item in the choice set. This means that each choice alternative is evaluated independently of the others as a share of total utility.

Regret Minimization

We often see consumers make decisions that minimize their regret at the possibility of having made a poor choice from all the alternatives that are available. In marketing analytics, this is called “regret minimization.”

The math of the regret model is more congruent with Behavioral Economics, where the context of the decision is the attractiveness of the current choice versus other available choices.

Utility Maximization & Regret Minimization

These two approaches deliver very different share predictions and implications for action. The utility maximization embedded in MNL only looks at one choice alternative at a time, while regret minimization looks at all choices and how they compare to one another.

Learn More

It can be difficult to decide which model best fits your choice data, but in4mation insights makes it easy. To learn more about the various choice modeling approaches and how they can obtain better results for decision-makers, please contact me today – Schedule a call or Send an email.


About Steve Cohen 

Steve Cohen is an award-winning entrepreneur, specializing in design of research, analysis of marketing data, and using marketing science tools to solve business problems. In 1984, Steve was the first commercial researcher to do choice-based conjoint and subsequently, he authored the first conference presentation and paper on MaxDiff. He is a co-author of the first academic paper on Latent Class Choice-based Conjoint and Menu-based Conjoint. Steve was won numerous awards for his research including: AMA “Parlin” Award, INFORMS Society for Marketing Science “Buck Weaver” Award, NextGen Marketing Research Disruptive Innovator Award, NY Marketing Research Council’s Hall of Fame, ESOMAR Goodyear Award for the Best International Research Paper, AMA Hardin Award for the Best Paper in Marketing Research Magazine, and Best Presentation at Sawtooth Software Conference. Read More