Target Table requirements

[Eduardo Robali1]

Good afternoon, 

We are in the early stages of starting to use Document Intelligence. While creating a use case, we noticed that there is a need to specify a target table to store the extracted values, such as: Account number, Service Address, Invoice date, etc. I have researched and watched some of the videos around Doc Intel, but have not been able to get an answer to my question. I want to know if for each use case a new target table needs to be created, and if so, this will be a custom table, which will most likely count against our custom table subscription, which varies by the contact we have signed with ServiceNow. Or is this type of table not taken into consideration against custom table subscription. 

[Loic1 · Accepted solution]

Hi @Eduardo Robali1,

To use the OOB solution, you indeed need to specify a Target Table. This table can be an existing table if the table has all the fields required to store the data. Or it could be a custom table, in that case, it will count against your custom table subscription.

The benefit of using a table is that it'll retain information in case you need to audit the data later or if you have workflows that span over multiple days/weeks. It is also used to trigger the extraction process and can be exposed to the end user via Record Producer, allowing for an end-to-end solution.

In case you only need the data to be temporarily extracted and available, one option would be to extend one of the exempt tables, such as sys_import_set_row. (FYI, the table subscription info is available here:  https://www.servicenow.com/content/dam/servicenow-assets/public/en-us/doc-type/legal/custom-table-gu...)

Alternatively, you can directly fetch the data from the di_extracted_value table and bypass using a staging table altogether. The "Retrieve Extrcated Values" Action in Flow Designer is doing something similar. This is more custom.

Hope that helps.

[Eduardo Robali1]

Thanks @Loic1 for your assistance.