Truace
Subscribe
68/100

Moderate evidence

Cite
More
Export CSVExport JSONEmailLinkedInX

Shopping chatbot's factual questions lifted buying intent; evocative ones hurt it, 335 tested

Study title: When and What Conversational AI Should Ask: Extending the Persuasion Knowledge Model to Explain User Engagement and Purchase Intention

Evidence type
Controlled or validated comparison
Sample
n = 335
Sources
1 sourcePeer-reviewed
Published
2 min read
Truvace record
TRV-2026-2563cite or check it

Study title: When and What Conversational AI Should Ask: Extending the Persuasion Knowledge Model to Explain User Engagement and Purchase Intention

Researchers built a recommendation chatbot using retrieval-augmented generation and tested it in an online experiment with 335 participants. They varied when the chatbot asked questions (during the conversation or as follow-up prompts) and what kind it asked (informational or experience-evoking).

Read the study abstract

Conversational AI is rapidly expanding in e-commerce, where chatbot questions critically shape users’ cognition and engagement. Yet, the effects of question timing and type on persuasion remain underexplored. Building on the persuasion knowledge model, this study examines how question timing (in-conversation vs. follow-up) and type (informational vs. experience-evoking) influence users’ knowledge structures, flow experience, and purchase intention. We developed a retrieval-augmented generation-based recommendati…

59Gain side

How the two sides scoreClose call

59Problem side

The gain

In an online experiment with 335 participants, informational chatbot questions asked during the conversation strengthened shoppers' topic knowledge, which positively predicted flow experience and purchase intention.

The problem

In the same experiment, experience-evoking questions asked during the conversation most strongly activated persuasion knowledge, which negatively predicted flow experience and purchase intention.

A fixed formula sets both scores from the evidence and its sources. Reader votes don’t move them.

What the scores mean

G is the gain side, out of 100. It goes up when the gain is specific, when it measurably helped people and when the sources behind it are strong.

P is the problem side, out of 100. It goes up when the problem is specific, when it measurably hurt people and when the sources behind it are strong.

In brief

Informational questions asked during the conversation, and experience-evoking questions asked as follow-ups, strengthened users' topic knowledge. Topic knowledge in turn positively predicted flow experience and purchase intention.

Experience-evoking questions asked during the conversation most strongly activated persuasion knowledge, meaning users' awareness of being persuaded. That awareness negatively predicted flow and purchase intention, so the same chatbot feature can build or erode buying interest depending on its timing.

Main points

  1. The study used a retrieval-augmented generation-based recommendation chatbot and an online experiment with 335 participants.
  2. It crossed question timing (in-conversation vs. follow-up) with question type (informational vs. experience-evoking).
  3. Topic knowledge positively predicted flow components and purchase intention.
  4. Persuasion knowledge negatively predicted flow components and purchase intention.
  5. Agent knowledge was activated under certain timing-type combinations but did not directly predict flow experience or purchase intention.

The full story

The work extends the persuasion knowledge model, which describes what people know about a topic, about persuasion tactics and about the agent trying to persuade them. It applies the model to conversational AI in e-commerce, where chatbot questions shape how users think and engage.

The effects depended on pairing timing with type. Informational questions worked best during the conversation, while experience-evoking questions strengthened topic knowledge only when delivered as follow-up prompts. Placed mid-conversation, experience-evoking questions triggered persuasion knowledge most strongly.

Agent knowledge, meaning users' understanding of the chatbot itself, varied across timing-type combinations. It did not directly predict flow experience or purchase intention.

What this doesn’t fix

The findings come from a single online experiment with 335 participants using one purpose-built recommendation chatbot, and measure purchase intention rather than actual purchases.

Sources

1 on file
  1. International Journal of Human–Computer Interactiondoi:10.1080/10447318.2026.2728933
    Peer-reviewed2026-09-11

Talk it over

Four questions to start onThis record found a gain and a problem in the same place. Which matters more to you, and who might answer differently?

Start with a question

Shopping chatbot's factual questions lifted buying intent; evocative ones hurt it, 335 testedhttps://truvace.com/item/when-and-what-conversational-ai-should-ask-extending-the-persuasion-knowledge-mo

  1. This record found a gain and a problem in the same place. Which matters more to you, and who might answer differently?

    The gain In an online experiment with 335 participants, informational chatbot questions asked during the conversation strengthened shoppers' topic knowledge, which positively predicted flow experience and purchase intention.

    The problem In the same experiment, experience-evoking questions asked during the conversation most strongly activated persuasion knowledge, which negatively predicted flow experience and purchase intention.

    Answer in the thread

  2. How do we know? What can this kind of evidence show, and what can’t it?

    The evidence This rests on a controlled comparison with a sample of 335, from International Journal of Human–Computer Interaction.

    Answer in the thread

  3. What does it leave open? What else would you want to know before acting on it?

    The record’s own caveat The findings come from a single online experiment with 335 participants using one purpose-built recommendation chatbot, and measure purchase intention rather than actual purchases.

    Answer in the thread

  4. Have you seen AI used this way where you study, work or live? Did it go the same way?

    Answer in the thread

The questions are the same on every record. The details under them are taken from this record as written.

What readers say

More from Business