Situation

We were engaged to assist our client with investigating internal employee records—the investigation aimed to ensure that an employer had paid the correct wages to its employees.

The legal team received over 10,000 employee timesheets with multiple data points that needed to be captured from each sheet. As an added complexity, the timesheets were among scanned hardcopy documents completed by hand. The client required a workflow to identify timesheets and extract relevant data to facilitate reconciliation.

Solution

Firstly, Sky Discovery developed an advanced search workflow using a combination of traditional and analytical search functions; we could identify which digitised documents were timesheets.

This workflow required minimal input from the legal team, except for quality assurance checks to ensure accuracy and effectiveness. After extracting the timesheets, we utilised Sky Discovery's Advanced OCR technology to generate text for the handwritten entries, which are inaccessible through traditional OCR methods.

The team utilised a Sky Discovery tool called Sky Extract to automatically extract timesheet data from designated locations on each timesheet. This extracted data was then imported into specific database fields for further analysis and use in the wage reconciliation process. The legal team focused on using the tool for quality assurance checks rather than manually extracting all information.

Impact

Our work in developing this custom, technologically powered workflow allowed our client to:

• Quickly and easily identify timesheet documents within a larger document set

• Obtain a more accurate set of extracted text from handwritten documents

• More accurately search using advanced features like Sky Extract; and

• Save hundreds of hours by automatically obtaining critical information from each timesheet.

As the process is not always accurate, the legal team still spent time ensuring the quality of the extraction. However, it required far less time than historical review methods.