Using generative AI to iteratively test product concepts
Published on
15 June 2023
Nik Samoylov image
Nik Samoylov
Founder

Want to learn how AI will take market research to the next level? In this presentation we discuss using generative AI to iteratively test product concepts, in particular Large Language Models.


Presented initially at Quant UX Con 2023, this presentation discusses the potential of Generative AI to iteratively test product concepts. The following three case studies are covered:

  • Using LLMs to probe respondents as to why they would recommend others to live in their city.
  • Using LLMs as the prompt for Gabor-Granger questions.
  • Building a respondent’s perfect cereal through multiple rounds of questioning from the LLM.

Using generative AI to iteratively test and refine product concepts

Originally presented at Quant UX Con 2023.

Do you have any feedback, or ideas on how you would use generative AI in your research? You are welcome to Book a call to discuss with our team.


Read these articles next:

Can consumers distinguish real and AI images?

Can consumers distinguish between real and AI images in 2025?

As AI technology advances, how confident are consumers in distinguishing between real and AI images, and does their confidence match their performance? Building on previous studies from 2023 and 2024, Conjointly examines these questions alongside evolving consumer attitudes towards AI-generated marketing content.

View article
Conversational versus open-ended questions surveys

Conversational versus open-ended questions surveys

Conjointly compared conversational surveys and typical open-ended questions to demonstrate when each approach works best and how to implement them effectively, using fast food consumption as the research topic.

View article
Conjointly launches Deep probe feature

Conjointly releases new study on consumer ability to distinguish real from AI-generated images in 2025

Conjointly, an all-in-one survey research platform, today published new research revealing that consumer ability to distinguish between real and AI-generated images has declined to chance levels, despite many consumers expressing confidence in their detection abilities.

View article