NCVO Civil Society Almanac
The National Council for Voluntary Organisations publishes an annual flagship report, the Civil Society Almanac (CSA). It is a respected source of data, trend information and insights into the voluntary sector.
NCVO wanted to streamline their process for assessing the changing sources of income for charities and changes in their expenditure patterns.Bricolage was brought in to help investigate the feasibility of automating the extraction of text and financial data from charities’ annual returns.
The project team tested open source Natural Language Processing (NLP) and data extraction tools on data samples provided by NCVO.
We also provided recommendations to NCVO for the key functional features that should be included in any system that they might be evaluating for purchase to automatically extract text and financial data from PDFs of annual financial returns.
The end result was a roadmap to assist NCVO as it moved away from inefficient manual extraction of data towards an automated system that saved time and money and extended the range of charities whose data could be included in the analysis behind the annual CSA report.
