Skip to main content
About HEC About HEC
Summer School Summer School
Faculty & Research Faculty & Research
Master’s programs Master’s programs
Bachelor Programs Bachelor Programs
MBA Programs MBA Programs
PhD Program PhD Program
Executive Education Executive Education
HEC Online HEC Online
About HEC
Overview Overview
Who
We Are
Who
We Are
Égalité des chances Égalité des chances
HEC Talents HEC Talents
International International
Sustainability Sustainability
Diversity
& Inclusion
Diversity
& Inclusion
The HEC
Foundation
The HEC
Foundation
Campus life Campus life
Activity Reports Activity Reports
Summer School
Youth Programs Youth Programs
Summer programs Summer programs
Online Programs Online Programs
Faculty & Research
Overview Overview
Faculty Directory Faculty Directory
Departments Departments
Centers Centers
Chairs Chairs
Grants Grants
Knowledge@HEC Knowledge@HEC
Master’s programs
Master in
Management
Master in
Management
Master's
Programs
Master's
Programs
Double Degree
Programs
Double Degree
Programs
Bachelor
Programs
Bachelor
Programs
Summer
Programs
Summer
Programs
Exchange
students
Exchange
students
Student
Life
Student
Life
Our
Difference
Our
Difference
Bachelor Programs
Overview Overview
Course content Course content
Admissions Admissions
Fees and Financing Fees and Financing
MBA Programs
MBA MBA
Executive MBA Executive MBA
TRIUM EMBA TRIUM EMBA
PhD Program
Overview Overview
HEC Difference HEC Difference
Program details Program details
Research areas Research areas
HEC Community HEC Community
Placement Placement
Job Market Job Market
Admissions Admissions
Financing Financing
FAQ FAQ
Executive Education
Home Home
About us About us
Management topics Management topics
Open Programs Open Programs
Custom Programs Custom Programs
Events/News Events/News
Contacts Contacts
HEC Online
Overview Overview
Executive programs Executive programs
MOOCs MOOCs
Summer Programs Summer Programs
Youth programs Youth programs
Article

Predicting Consumer Choices With New Eye-Tracking Data

Marketing
Published on:

Exploring consumer preferences is an essential step in predicting consumer purchase decisions for companies. And yet, market research often wrongly assumes that consumers rationally base their decisions on all available information. A new model addresses this flaw by using eye-tracking data to measure how consumers collect information and improve the understanding and prediction of consumer choices.

Using eye-tracking to understand and predict consumer choices - Cathy Yang @Fotolia-Sergey Nivens

“Deciding which movie to see [in a twenty-screen multiplex] may be so hard that you end up staying home and watching TV,” writes US psychologist Barry Schwartz in a companion essay to his highly influential book on choice overload, The Paradox of Choice. That's because, he writes, “choosing is work: it takes time, attention and effort.” And our resources – in time, abilities, etc. – are limited, meaning that the rationality of our decision-making is limited too, an effect that economists have named “bounded rationality”. As a consequence, individuals facing a complex choice (dozens of movies and show times, or an entire supermarket aisle of chocolate bars) will sometimes give up on sifting through all the product information and choose the best option given the information they have collected up to a point. Accounting for this consumer decision-making process usually helps improve the prediction of individual purchase decision. However, the major challenge is that this decision process is hard to observe.

Recognizing the limitations of existing statistical models 

Since the early 1980s, business has relied heavily on choice-based questionnaires to collect consumer preference data, using statistical models to recover consumer preferences. Without observations of the decision process, existing statistical models usually assume that consumers collect all choice-relevant information before making a decision. Since the way in which consumers search for choice-relevant information reveals their preferences, a statistical model with wrong assumptions about these information acquisition processes could bias marketing managers' estimations of consumer preferences and make prediction of purchases less accurate. How then can market research take these limitations into account? A team of researchers has come up with a statistical model that does just that, and should help marketing practitioners improve the accuracy of measuring consumer preference and therefore the prediction of consumer choices with shorter questionnaires. 

Modeling the information acquisition process 

The researchers observed that people don't process all relevant information when making choices, such as, for example, when choosing a laptop. The area of the questionnaire on their online platform that consumers attend to reveals their preferences: “Maybe a teenage girl will only attend to the color and size of the screen whereas a teenage boy will also care about the speed of the processor to play games,” explains Cathy Yang, one of the researchers. That's why she and her co-authors developed a model to capture information acquisition (finding out about the relevant information for purchasing a laptop) as a result of forward-looking utility maximization, in which the utility a consumer derives comes not only from the chosen product but also from the information acquisition process itself (is it worth reading about and assessing x specifications).

 

Not all consumers will attend to the same features.

 

Eye-tracking as a measure of attention and information acquisition

Recent marketing studies have relied on eye-tracking data as a measure of attention allocation for many purposes, from branding to search effectiveness and brand display on supermarket shelves, and the authors believe it will become more and more widespread, if not systematic, as a way of fine-tuning the understanding of how consumers make purchase decisions. As Cathy Yang recalls, in the late 1980s and early 1990s, marketers relied on mouse-tracking to understand the information acquisition process: the subject's information acquisition processed was captured by tracking the mouse clicks necessary to open boxes which contained relevant information presented on a computer screen. “But it was an intrusive, non-natural process, whereas an eye-tracker looks just like a regular screen,” she adds. In this study, the authors leverage eye-tracking data as a direct measure of attention and information acquisition. Eye-tracking data is one way of inferring the decision process and signaling which features of a product are more important to different users. For instance, when presented information about the same laptop, one consumer might devote 80% of his or her attention to the screen attributes, 15% to hard disk capacity and only 5% to the rest. “Not all consumers will attend to the same features,” says Cathy Yang. Managers can gain insight into consumers’ heterogeneous preferences from the eye-tracking data. 

It is technical to collect eye-tracking data, but, today, commercial solutions are available for collection of eye-tracking data in an online environment using webcams. “We expect such solutions to be increasingly common as large companies such as Facebook acquire such capabilities […] and with the development of open-source solutions,” write the authors.

Applications for Managers

Image - Social Networks
This model gives a good understanding of consumer preference by modeling information acquisition process. More importantly, it enables marketing practitioners to better predict consumer choices with shorter questionnaires: this method requires 1/3 of the questions necessary with a traditional model in order to achieve the same choice prediction accuracy if applying a traditional model. As Cathy Yang points out, if marketing practitioners want to present questions about 6 “attributes” (or features, e.g. the screen size of a laptop) of a product, with 4 “levels” each (13 inches, 15 inches, and so on), the questionnaire is already quite long, adding: “The problem is that people pay less attention at the end, whereas if you make it short, you have higher-quality data to predict consumers’ individual choice” – and fewer chances that the product will flop because of flawed market research.

Methodology

methodology
For their survey, the researchers recruited 70 participants from a large European university and had them answer 20 questions about their preferences concerning six attributes for a Dell laptop computer (price, hard drive capacity, and so on), using an online platform developed by the authors. They complemented the results with eye-tracking data (fixation of eyesight on a specific location and saccades, or movements between fixations) collected on the participants, also in the university's behavioral laboratory.
Based on an interview with Cathy Liu Yang on her article “A Bounded Rationality Model of Information Search and Choice in Preference Measurement”, co-authored with Olivier Toubia and Martin G. De Jong (Journal of Marketing Research, April 2015).