Purpose This study investigates how investor sentiment influences equity-based crowdfunding campaigns outcomes, with a particular focus on sentiment related to specific industries. The motivation stems from the need to understand the behavioural drivers behind crowdfunding outcomes, especially in a context where both private and public funding coexist. Study Design We employed a machine learning approach, utilizing the PolBERT model to analyse investor sentiment from online forum discussions in Poland. The data consisted of sentiment expressed across various industry-specific discussions, which were analysed to assess their impact on crowdfunding campaign outcomes. Findings The study reveals that while positive investor industry sentiment enhances the capital raised for IPO-targeted companies, it negatively affects overall campaign success and the number of investors, particularly in non-IPO-targeted firms. These findings suggest a complex relationship between sentiment and campaign outcomes, where higher sentiment can sometimes deter participation in certain cases. Contributions This research introduces a novel category of behavioural success drivers in the context of equity crowdfunding, proposing a new classification of meta-drivers of crowdfunding success. Methodologically, the study demonstrates the efficacy of fine-tuned BERT-class models in processing language, offering new insights into the investor sentiment analysis. Implications The results have significant implications for both researchers and practitioners. For researchers, the study expands the understanding of behavioural finance in crowdfunding, while for practitioners, it provides strategic insights into how investor sentiment can influence campaign timing, structure, and expectations. The findings also offer value for policymakers by shedding light on investor behaviour in equity-based crowdfunding markets.