Market research and knowledge using Generative AI: the power of Large Language Models

Generative artificial intelligence (GAI) is rapidly transforming the marketing industry with its capabilities in data analysis, personalisation, strategic optimisation, and content generation, providing powerful tools for businesses aiming to establish or maintain a strong brand position. This research examines the application of Large Language Models (LLMs) for market research, enabling marketers to create insights that capture the complexities of market behaviour and customer psychology, helping them to connect with their target audiences more meaningfully and effectively. Particularly, it aims to evaluate the extent to which GAI can replicate traditional market research. We conducted a comprehensive survey on beer consumption in Spain, covering all conversion funnel stages, from Brand Awareness to Purchase. The study was replicated using four prominent LLMs: ChatGPT (OpenAI), Gemini (Google), Claude (Anthropic) and LlaMa (Meta), and the results of these LLMs were compared with those of the traditional survey using a collection of statistical methods. Our results show that LLMs are valuable for market research, offering significant insights as reliable proxies. This represents a competitive advantage by making studies of this kind more accessible and cost-effective, benefitting companies of all sizes. However, LLMs cannot fully replicate traditional methods and present result variability, introducing risks in decision-making because of potential errors in data generation that are complex to estimate without benchmarking.