'Document Extraction' Actions


The Document Extraction component automatically extracts the relevant data from files attached to incoming emails so that this data can be used in further processing of the work item, saving your agents time and effort. This also means that documents such as PDFs can be scanned and used both to start Cases in Enate and to form part of the ongoing process's activities.

When a Document Extraction Action runs for a Case, documents attached to the Case can be submitted to your desired technology for scanning and the processed output files will be returned and automatically attached to the Case.

If at any point the technology you're using is not confident enough of the results, based on a confidence threshold that you can set, Enate will instantly transfer the work to an agent in Work Manager to look over and verify, giving you that 'human in the loop' support.

This component can be switched on by your admin in the Marketplace section of Enate Builder.

Check out this video to find out more:

How it works at runtime

When the Case is run in Work Manager, relevant data from files attached to incoming emails for it will be automatically analyzed and extracted.

If the technology you're using is confident enough about its data extraction results, this Action won't even need to be seen by a human user, it will simply be completed automatically and the Case will move on to the next Action. The completed data extraction Action can still be viewed if you click on it, but it won't need to be handed over to a human user for involvement.

However, if the extraction technology is less confident in its data extraction results, the Action will be handed over to a human user when you next hit 'pull from Queue' in their home page, to pick up and look over. When an agent opens the Action, you'll see that it's been given to them because some further checks are required.

To do this, you just needs to click on 'Verify Now' and scroll to the 'validation station' screen in the Action, which shows the scanned document image and the resulting extracted table of data values. This lets you see where those lower confidence levels are highlighted, review them and make any necessary corrections manually. This can viewed in-situ, or expanded out to a popup to display full screen.

Every time this is done, the technology will learn and get a little bit better at its data extraction suggestions. If you notice that the technology is regularly getting its suggestions wrong, speak to your admin team about modifying the confidence threshold.

Once you are happy that the extracted data is as desired, you can click to mark the Action as completed successfully.

Last updated