Synthetic Consumer Insight Generation with Large Language Models

arXiv cs.AI Papers

Summary

This research examines whether LLMs can generate synthetic consumer data for projective techniques, comparing human and LLM responses on city tourism perceptions and finding substantial overlap but differences in style and diversity.

arXiv:2607.05761v1 Announce Type: new Abstract: Modern data-driven marketing relies on large amounts of consumer data, yet collecting such data can be costly, time-consuming, and difficult to scale. This research examines whether large language models (LLMs) can be used to generate synthetic consumer data for projective techniques, a set of methods designed to elicit consumer associations, emotions, wants, and needs. We test LLM-generated responses across multiple projective tasks, LLMs, prompting strategies, and temperature settings, and compare them with human responses from a primary research study on perceptions of city tourism destinations. Human and LLM responses were analyzed using linguistic measures, diversity and concentration metrics, topic models, and top-term analyses. The results show substantial overlap between human and LLM responses in broad topics and associations, but also important differences in style, linguistic structure, and the way diversity is generated. Recommendations are given on how to best utilize LLMs for generating synthetic consumer data, how model and prompt choices shape response quality, and on recognizing the limitations of LLM synthetic consumer data generation.
Original Article
View Cached Full Text

Cached at: 07/08/26, 04:38 AM

# Synthetic Consumer Insight Generation with Large Language Models
Source: [https://arxiv.org/abs/2607.05761](https://arxiv.org/abs/2607.05761)
[View PDF](https://arxiv.org/pdf/2607.05761)

> Abstract:Modern data\-driven marketing relies on large amounts of consumer data, yet collecting such data can be costly, time\-consuming, and difficult to scale\. This research examines whether large language models \(LLMs\) can be used to generate synthetic consumer data for projective techniques, a set of methods designed to elicit consumer associations, emotions, wants, and needs\. We test LLM\-generated responses across multiple projective tasks, LLMs, prompting strategies, and temperature settings, and compare them with human responses from a primary research study on perceptions of city tourism destinations\. Human and LLM responses were analyzed using linguistic measures, diversity and concentration metrics, topic models, and top\-term analyses\. The results show substantial overlap between human and LLM responses in broad topics and associations, but also important differences in style, linguistic structure, and the way diversity is generated\. Recommendations are given on how to best utilize LLMs for generating synthetic consumer data, how model and prompt choices shape response quality, and on recognizing the limitations of LLM synthetic consumer data generation\.

## Submission history

From: Stephen L\. France \[[view email](https://arxiv.org/show-email/827ad16a/2607.05761)\] **\[v1\]**Tue, 7 Jul 2026 02:38:41 UTC \(985 KB\)

Similar Articles

Evaluating LLMs as Human Surrogates in Controlled Experiments

arXiv cs.CL

This paper evaluates whether off-the-shelf LLMs can reliably simulate human responses in controlled behavioral experiments by comparing LLM-generated data with human survey responses on accuracy perception. The findings show that while LLMs capture directional effects and aggregate belief-updating patterns, they do not consistently match human-scale effect magnitudes, clarifying when synthetic LLM data can serve as behavioral proxies.

When Synthetic Users Fail: A Cross-Domain Benchmark of LLM-Simulated Human Survey Responses

arXiv cs.CL

This paper benchmarks LLM-simulated human survey responses across two large-scale datasets, finding that no model beats simple baselines at the individual level and that models systematically over-determine demographics, distorting segment differences. The failures persist across model scales and families, raising concerns about using synthetic users for decision support.