[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82991-en":3,"doc-seo-82991-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},82991,7971461740886,"Theodore","https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2",8,"Research & Report","Synthetic Consumer Insight Generation with Large Language Models","Modern data-driven marketing depends on extensive consumer data, but collecting it is costly, time-consuming, and difficult to scale. This research investigates whether large language models (LLMs) can generate synthetic consumer data for projective techniques that capture associations, emotions, wants, and needs. Responses are tested across multiple projective tasks, prompting strategies, and temperature settings, then compared with human responses using linguistic, diversity/concentration, topic-model, and top-term analyses. Results highlight overlap in broad topics while revealing stylistic and diversity-generation differences, yielding practical recommendations and limitations.","Synthetic Consumer Insight Generation with Large Language Models  \nStephen L. France (Corresponding Author)  \nMississippi State University Mailstop 9582, Mississippi State, MS 39762 Phone: +1 662-325-1630 , Fax: +1-662-325-7012  \n[sfrance@business.msstate.edu](sfrance@business.msstate.edu)  \nORCID iD: 0000-0002-7226-2750  \nPia A. Albinsson  \nBeroth Professor of Marketing  \nAppalachian State University  \nASU Box 32090  \nBoone, NC 28608-2090  \n[albinssonpa@appstate.edu](albinssonpa@appstate.edu)  \nORCID iD:0000-0002-8591-5191  \nSynthetic Consumer Insight Generation with Large Language Models  \nABSTRACT  \nModern 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 topterm 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.  \nKeywords: Synthetic Data Generation, Large Language Models, Agentic Information Systems, Topic Modeling, Marketing/IS Interface, Projective Techniques  \nINTRODUCTION  \nGenerative AI (artificial intelligence) using LLMs (large language models) is increasingly used to generate new datasets and augment existing datasets in applications as diverse as generating  \ntraining data to improve machine learning performance (e.g., Li et al., 2023; Tang et al., 2023), generating health data for clinical discovery and disease prediction (Kang et al., 2024 ; Smolyak et al., 2024), and generating synthetic personality personas that can be used to simulate different types of users, customers, or decision makers in applied research tasks (Ge et al., 2024) . These data “generation” capabilities can easily be applied to consumer behavior settings, where marketers require consumer data to help understand consumers’ preferences and motivations (e.g., Basu et al., 2023), but where gathering data can be expensive and time consuming (Market Research Society, 2024) . A key aim of synthetic data generation is to produce output that has similar data, including key data points and word distributions, to human responses (Rossi et al., 2024) . From an information systems (IS) perspective, LLM-generated synthetic consumer data can be understood as a form of AI-enabled information generation, because generative AI systems produce new, meaningful content by drawing on patterns learned from large corpora that may include consumer reviews, social media posts, blogs, and other forms of consumergenerated text (Feuerriegel et al., 2024; Brand et al., 2023) .  \nThe focus of this research is first on showing how LLMs can be used to generate synthetic consumer data for creative and interpretive marketing research tasks, We use projective techniques as the methodological testbed because they elicit open-ended consumer associations rather than direct ratings or choices, making them useful for evaluating whether LLMs can generate richer forms of consumer insight. Tourism destinations provide an appropriate empirical context because they evoke factual knowledge, emotions, cultural imagery, and symbolic associations, all of ","cbCailsMpx26GqXR","https://ap.wps.com/l/cbCailsMpx26GqXR","pdf",1008794,2,1,48,"English","en",105,"# Abstract\n# Introduction\n## Research questions\n# Background","[{\"question\":\"How do the study’s projective techniques relate to consumer insight generation?\",\"answer\":\"Projective techniques elicit open-ended consumer associations, emotions, wants, and needs. This makes them suited to evaluating whether LLMs can produce richer, interpretive insight beyond direct ratings or choices.\"},{\"question\":\"What variables are tested to evaluate LLM-generated synthetic consumer responses?\",\"answer\":\"The study tests LLM-generated responses across multiple projective tasks, different LLMs, prompting strategies, seed examples, and temperature settings to examine how these choices affect output quality.\"},{\"question\":\"How are LLM outputs compared with human responses?\",\"answer\":\"Human and LLM responses are analyzed using linguistic measures, diversity and concentration metrics, topic models, and top-term analyses to assess similarity, diversity, and interpretive richness.\"}]",1784184504,121,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"synthetic-consumer-insight-generation-with-large-language-models","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/synthetic-consumer-insight-generation-with-large-language-models/82991/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How do the study’s projective techniques relate to consumer insight generation?","Question",{"text":75,"@type":76},"Projective techniques elicit open-ended consumer associations, emotions, wants, and needs. This makes them suited to evaluating whether LLMs can produce richer, interpretive insight beyond direct ratings or choices.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What variables are tested to evaluate LLM-generated synthetic consumer responses?",{"text":80,"@type":76},"The study tests LLM-generated responses across multiple projective tasks, different LLMs, prompting strategies, seed examples, and temperature settings to examine how these choices affect output quality.",{"name":82,"@type":73,"acceptedAnswer":83},"How are LLM outputs compared with human responses?",{"text":84,"@type":76},"Human and LLM responses are analyzed using linguistic measures, diversity and concentration metrics, topic models, and top-term analyses to assess similarity, diversity, and interpretive richness.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]