423. IJRM. Passive Power.

 

August 6, 2026

Listen to full episode :

423. IJRM. Passive Power.
Gabriella Mirabelli

Can the grammar of a complaint predict whether a customer will escalate?

In this episode, Gabriella Mirabelli speaks with Rod Duclos, Associate Professor of Marketing at Ivey Business School at Western University, about research showing that passive voice in customer complaint text reliably predicts escalation risk. The paper, co-authored with Amir Sepehri and Jonah Berger, was published online in April 2026 in the International Journal of Research in Marketing (IJRM).

What the research found

Customers who write complaints in passive voice ("the wrong PIN was entered") are more likely to escalate than those writing in active voice ("I entered the wrong PIN"). The passive construction signals that the customer assigns fault to the service provider rather than to themselves. The study analyzed over 160,000 complaints filed with the Consumer Financial Protection Bureau between 2015 and 2017, then replicated the finding across five controlled experiments.

Why grammar outperforms content analysis alone

Explicit complaint content often stops short of naming the company as responsible. Grammatical voice captures the blame attribution that content misses. The effect holds after controlling for complaint length, emotional tone, month, weekday, state, and dozens of other variables.

What practitioners should do with it

Duclos recommends a three-step approach: audit historical complaint data using the passive-voice measurement tool available free on co-author Amir Sepehri’swebpage; use passive voice as one feature in a larger triage model, not as a standalone decision rule; and test interventions (response speed, escalation to senior reps, compensation timing) before scaling.

The AI confound

Most AI writing assistants default to active voice. If customers increasingly draft complaints through tools like Grammarly or ChatGPT, the syntactic signal may erode. Duclos expects blame attribution to migrate into content signals instead, including causal language, fairness claims, and explicit demands.

Click here for the free passivevoice tool from Amir Sepehri

Click here for the Research Article

The Up Next podcast’s access to this content is courtesy of the International Journal of Research in Marketing, an international, double-blind peer -reviewed journal for marketing academics and practitioners. IJRM aims to contribute to the marketing discipline by providing high-quality, original research which advances marketing knowledge and techniques. As marketers increasingly draw on diverse and sophisticated methods, IJRM‘s target audience is comprised of marketing scholars, practitioners (e.g., marketing research and consulting professionals) and policymakers.

IJRM aims to be at the forefront of the marketing field with a particular emphasis on bringing timely ideas to market. The journal embraces innovative research with the potential to spur future research and influence practice. Hence, it welcomes contributions in various aspects of marketing. The editors, while accepting a wide array of scholarly contributions from different disciplinary approaches, especially encourage research that is novel, visionary or path breaking.

Interviewee

Rod Duclos is a behavioral psychologist. His research aims to articulate, predict, and shape human decision-making and behavior, all in the pursuit of consumer and societal welfare. Particular issues of interest include: the psychology of financial decision-making and risk-taking; how hormones impact willingness to pay; prosocial behavior; and judicial decision-making. His findings have appeared in premier marketing and psychology journals such as the Journal of Consumer Research, the Journal of Consumer Psychology, and the International Journal of Research in Marketing. His work has also been covered over 100 times in the popular press (e.g., TIME Magazine, CBS, NBC News, US News & World Report, Harvard Business Review). He owes his PhD to the University of North Carolina at Chapel Hill and Duke University.

Prior to joining the Ivey Business School, he taught at the Hong Kong University of Science & Technology (HKUST) and the University of North Carolina at Chapel Hill (Kenan-Flagler Business School). His lectures range from intro courses on Marketing Principles to advanced courses on Global Marketing and Strategic Brand Management.

In addition to research and teaching, Professor Duclos consults for private, governmental, and nonprofit organizations. His clients range from the Americas to Europe to Asia.

Co-Authors

Amir Sepehri is an Assistant Professor of Marketing at ESSEC Business School. His research interests are at the intersection of psycholinguistics and human-AI interaction. 

Johnah Berger is a Marketing professor at the Wharton school. His research revolves around language and social influence. 

Follow and learn more:

International Journal of Research in Marketing

  • Latest Content here

  • Follow on LinkedIn here

Rod Duclos

 
Next
Next

422. Joshua Lifrak. Win Today.