Journal of the Association for Consumer Research (2026) — language & word-of-mouth length

Summary:

This study asks what keeps people talking longer about products, brands, services, or experiences. It focuses on language in real back-and-forth conversations — especially how concrete people are and what kinds of questions they ask.

Published in the Journal of the Association for Consumer Research, the paper (Reihane Boghrati and Jonah Berger) tests whether certain language patterns extend word-of-mouth conversations, since longer discussions tend to be more persuasive.

Method

• Study 1: 192 participants (96 pairs) talked for 10 minutes about prelisted topics → 583 topic discussions, ~5,500 turns.
• Study 2: 222 participants had freer 10-minute conversations → 125 conversations, 745 topic discussions, 16,000+ turns.

Researchers transcribed the talks, scored each turn for concreteness, detected questions with a trained model, had assistants label questions as broad vs. narrow, and used turn-by-turn hazard models to predict when topics ended.

Findings

• Longer discussions were rated more persuasive (and in Study 1, more enjoyable).
• Concrete language kept topics going longer than vague wording.
• Questions helped, but mainly narrow questions (specific follow-ups). Broad questions didn’t reliably extend the topic.

Results held with extensive controls (speaker, relationship, interest, category, order, word count, complexity, emotionality, and other dialogue moves) and replicated in the more natural Study 2.

Caveats
It’s observational (not a language-manipulation experiment), the sample skewed young and female, and question-type coding mixed automation with human judgment. Some Study 1 sessions were on Zoom, though results were similar.