Utilizing Large Language Models to Synthesize Product Desirability Datasets
John D. Hastings, Sherri Weitl-Harms, Joseph Doty, Zachary J. Myers, Warren Thompson
TL;DR
The paper demonstrates that a cost-efficient LLM (gpt-4o-mini) can generate large synthetic PDT datasets via three methods (Word+Review, Review+Word, Supply-Word), achieving strong sentiment alignment (Pearson r ≈ 0.93–0.97) and notable text diversity, albeit with biases toward positive sentiments. It systematically analyzes alignment, diversity, and generation costs, showing Trade-offs: Supply-Word offers higher diversity at higher cost; Word+Review tends to be cheaper with slightly lower diversity. The work highlights practical viability for scalable PDT data production in low-resource scenarios, while addressing ethical considerations and biases. The results support the use of synthetic PDT data to augment limited real-world data for UX research and product design.
Abstract
This research explores the application of large language models (LLMs) to generate synthetic datasets for Product Desirability Toolkit (PDT) testing, a key component in evaluating user sentiment and product experience. Utilizing gpt-4o-mini, a cost-effective alternative to larger commercial LLMs, three methods, Word+Review, Review+Word, and Supply-Word, were each used to synthesize 1000 product reviews. The generated datasets were assessed for sentiment alignment, textual diversity, and data generation cost. Results demonstrated high sentiment alignment across all methods, with Pearson correlations ranging from 0.93 to 0.97. Supply-Word exhibited the highest diversity and coverage of PDT terms, although with increased generation costs. Despite minor biases toward positive sentiments, in situations with limited test data, LLM-generated synthetic data offers significant advantages, including scalability, cost savings, and flexibility in dataset production.