Automated Reporting of Corporate Sustainability Data – an Analog of Automated Internet-based Sustainability Reporting Using XML

Abstract

Extracting information from corporate sustainability reports for analysis is a labor-intensive process fraught with risks of misclassifying or missing relevant material. The use of a GRI index helps to some extent, but the main problem is a lack of standardization of language that would make automated searching feasible. The need for such standardization would be entirely obviated if every piece of data in the report were tagged in some machine-readable way, such as XML. This paper discusses a parallel process in which all data generated by the Roberts Environmental Center reside in specific database fields and are extracted automatically on demand to meet the needs of web pages and reports. The data for the web pages are retrieved in XML-tagged format for the purposes of driving web page graphics. These data are embedded in the web pages and could be processed by end users, if they wished, to create tables or other output in addition to the graphs. Exactly the same database-querying techniques could return all the data about a specific report in our database in a single XML document. Such an approach might be useful for companies wishing to tag their sustainability reports with XML. If all of the data for the report, including text blocks, were contained in a database (as is the case for all Roberts Center reports), the XML extraction routine would need to be written only once, and new data would automatically be available to web pages parsing the XML. Analyzing the content of such reports would be greatly simplified.