This repository features the document data and API of the Sustainabity Reporting Navigator. The large bulk of the initial reports in our database comes from the generous contribution by Arianna Pisciella, Gaia Melloni, Bianca Minuth, and Paul Pronobis and is supplemented by the ongoing data collection of the SRN team. Our objective is to develop this repository into a collaborative data platform that provides extensive coverage of sustainability-related documents published by European publicly-listed firms.

Currently our data covers 11,885 documents from 888 firms spanning 20 countries and data from the time period 2010 to 2023. Further information on the covered firm-years can be assessed from the table below.

We try to collect all documents that contain relevant sustainability information. This includes but is not limited to annual and sustainability reports (AR and SR). For some firms it also includes additional reports like integreated reports (IR), Carbon Diclosure Project data (CDP), and other reporting formats. All materials are referred to by URL links and are provided as is. Neither the team of the SRN nor the maintainers of this data repository claim any ownership of or legal rights to the provided data.

We provide an easy to use API for data download. See the file srn_docs_api.py in this repo for pointers. We are working on making this process even easier to handle over the next weeks. Stay tuned!

Thank you for asking! We very much appreciate you giving credit to our data collection efforts. We will be using a concept DOI and versioned DOIs from Zenodo in the future. Currently, please cite the data as follows (in alphabetical order):

Donau, Charlotte-Louise, Fikir Worku Edossa, Joachim Gassen, Gaia Melloni, Inga Meringdal, Bianca Minuth, Arianna Piscella, Paul Pronobis and Victor Wagner (2023): SRN Document Database, https://github.com/trr266/srn_docs.

What a great question! We really hope that this repo will be used to discuss data quality issues as well as methods to use the data in research projects. Again, as a first teaser, take a look at extraxt_text_from_docs.py that features a method to extract page-wise text data from our documents.

Thank you! Please open an issue on GitHub, mentioning the document id and describing the issue that you encountered.

Thank you and yes, of course! We would be very happy about people that provide historical data and/or are willing to maintain our data going forward on a country or index basis. So, if you would like to help, please reach out to Victor Wagner.