Beyond compliance: a textual analysis of drivers of sustainability reporting in Italian SMEs

Summary of Background Data Sustainability reporting has historically been the field of large companies. However, in recent years, an increasing number of SMEs (small and medium-sized enterprises) have shown interest and in already implemented, sustainability reporting tools [3]. This trend is partly driven by the growing number of regulations and government initiatives that make sustainability reporting mandatory for SMEs. This study examines the sustainability reports of 15 Italian companies, drafted between 2020 and 2023, selected from the top 30 companies ranked highest in ESG performance according to the annual evaluation conducted by the Standard Ethics Agency. These firms operate in sectors such as agri-food, processed food, and consumer goods. By employing textual analysis, the study identifies key themes and examines how organizational characteristics — such as market presence (national vs. multinational), industry affiliation, and financial metrics (e.g., return on equity, solvency ratio) — influence their engagement in sustainable activities. Objectives Our research aims to identify recurring topics in corporate sustainability reports, analyze the impact of company attributes on topic prevalence, and explore linguistic differences using advanced textual analysis techniques. Methods The dataset comprises sustainability reports segmented into sentences using Python scripts. Text cleaning was performed to remove meaningless terms and company-specific identifiers. The Structural Topic Model [2] was implemented to identify prevalent themes, while covariates such as market presence, sector, and financial metrics (ROE and solvency ratio) were integrated to assess their impact on topic prevalence. A pre-determined optimal number of 15 topics was selected based on established metrics. Additionally, a Keyness Analysis [1] was conducted to detect linguistic differences among company categories, focusing on performance, size, and sector. Results The STM identified 15 distinct topics, including global sales, sustainability governance, carbon emissions reporting, sustainable sourcing, human rights, and healthy nutrition. “Top Performers” emphasized sustainability governance and innovation, whereas “Vulnerable” companies focused more on operational challenges. Sector-specific themes emerged, with agribusiness prioritizing sustainable sourcing, while processed foods highlighted consumer health. Keyness Analysis revealed linguistic differences: smaller companies used localized terminology, while multinationals adopted globalized narratives. Sector- specific language further distinguished sustainability focuses, emphasizing sustainable sourcing and marine conservation in agribusiness and nutrition in processed foods. Discussion/Conclusions This study underscores how organizational and financial characteristics shape sustainability narratives. Financial stability, market presence, and industry-specific challenges significantly influence the themes and language of reports. The findings provide actionable insights for enhancing transparency, aligning corporate reporting with global sustainability standards, and improving stakeholder engagement. References 1. Gabrielatos, C.: Keyness analysis: Nature, metrics and techniques. In: Taylor, C. and Marchi, A. (eds) Corpus Approaches to Discourse: A Critical Review , chapter 11. Routledge, xford, (2018). 2. Roberts, M. E., Stewart, B. M., Tingley, D., Lucas, C., Leder-Luis, J., Gadarian, S. K., Albertson, B., and Rand, D. G. Structural topic models for open-ended survey responses. AJPS, 58(4), pp.1064–1082, (2014). doi: 10.1111/ajps.12103. 3. Stolowy, H., Paugam, L.: Sustainability reporting: Is convergence possible? Accounting in Europe, 20(2), pp. 139–165 (2023).