Automating the extraction of unstructured data

KPMG LLP’s proprietary technology uses emerging technology, including optical character recognition, to facilitate the extraction of relevant data from unstructured documents, such as bank confirmation PDFs. It can also identify matching data across multiple documents, ultimately resulting in efficient, high-quality reconciliation that is easy to trace back to source documents. 

KPMG is focused on a more automated, data-driven audit to deliver quality and efficiency to our clients. Automating our response to risk is integral to that vision. Enter our proprietary document data extraction and validation tool. 

In Action: For the last few years, a large financial services client and the KPMG team have pushed to implement new technology and innovation throughout the global audit engagement. Any time saved can help the KPMG team increase their focus on higher-risk areas, so they embraced this technology as soon as it became available.

“I kept thinking it was too good to be true,” one team member noted, “especially because the tool is so easy to use.”

The outcome? Another successful technology implementation to share with the client. By automating previously manual processes like these, KPMG teams can redistribute time and focus from low-judgment work to areas of increased risk – thereby optimizing time spent in relevant areas – overall, a better audit experience.