
The IUCN Red List of Threatened Species is a valuable resource for global conservation, widely used by researchers and policymakers. While its scale and coverage are impressive, it comes with only a limited resolution at finer spatial scales. Globally, least concerned species may still be declining or even extinct within certain parts of their range. National and regional red lists, produced at various geographical and taxonomic scales, provide this missing local perspective, yet despite their relevance, they remain largely overlooked in large-scale analyses. This is mainly due to poor accessibility of the documents, heterogeneity in formats, and the lack of a unified, digital, and standardised structure. Here we introduce RegRed, which addresses this gap by building a comprehensive, spatially and taxonomically standardised, per-species geodatabase of national and regional threat assessments, openly accessible to support research, policy, and conservation action. In the first phase, we conducted a systematic global search and compilation of red list sources, identifying over 2,000 documents from 172 countries (~90% of the world’s countries), covering threat assessments across 487 broad taxonomic groups. In the second phase, we focused on data digitisation, beginning with tetrapods, using a standardised extraction protocol and a semi-automated workflow that incorporates large language models (LLMs). We will present a teaser of RegRed’s first release, which as of June 2026 spans more than 64 countries, 16,000 species, and 800 sources.