Epicor IDC (DocStar) Class Verification

We are using IDC/ECM for AP Invoice Entry and are getting to the point where the longest part of the process is splitting and merging documents or attachments. It is my understanding that Class Verification is supposed to help with this. However, we did not initially set this up.

I would like to hear from those of you who use class verification for their AP Invoice entry process. Does it work well? If so, is it easy to set up and maintain the model? How do we do it? Would love to hear any advice or feedback you may have.

I think IDC added some things in 9.3 to address these without going to classification. I have not used them, but they are on my list to deal with vendors that send us multipage invoice pdfs.

This is the documentation area
https://epicweb.epicor.com/products/epicor-ecm/documentation
image
image

3 Likes

Building the Classification Model can be a lot of up front work, but it will automate the Class Verification step so that it identifies attachments and marks them as such. It can also split the batch into documents.

@gpayne mentioned the new features in IDC 9.3+ that allow for some of this, but these are part of the Document Form Definition. What this means is that the batch has already passed the point where you can mark pages as attachments and not be charged for them. If this isn’t an issue, then one of the other features does allow you to split a batch into respective documents upon finding a change in a field you specify. This means you can set it so that if it finds a new invoice number, it will split at that point and create a new document. That will continue through the rest of the batch, splitting into different documents whenever it finds a new invoice number.

2 Likes

There is a misc parameter (DocSeparator=InvoiceNum) that can be used in version to split document, it can’t classify pages as scannable or non scannable but it does help with the splitting task by seeing a change in the invoicenum and split to a new document as per @vleveris

1 Like

@vleveris @gpayne @swilliasc111

Thank you all for pointing me to the DocSeparator functionality in 9.3. I’ve enabled this and it is working, for the most part…

What I’ve found is that it is splitting documents every time it finds an invoice number. However, some of our vendors have the invoice number on every page, and it will split a multiple page invoice into multiple batches.

What I’m wondering is if this can be adjusted to only split documents upon finding a UNIQUE invoice number.

Any thoughts?

My understanding of the new feature is that it should only be splitting into different documents upon finding a different invoice number. I haven’t had much of an opportunity to explore this new feature, but I know my colleague @swilliasc111 did some work using it.

I would see if the InvoiceNum field links on a supplier document with no docseparator variable turned on.

My testing finds that if a document has the same invoice number on each page, the docseparator will only split when a new invoicenum is found. I have tested this functionality maybe 3 times.

Here is some information from the Ancora Software website.

The Document Separator misc parameter in DFD allows you to trigger a document break based on the field(s) it is configured for, in response to user training in Data Verify.

  1. To enable the Document Separator misc parameter, click the triple dot link for a field in the Misc. parameters column to open the Misc parameters for that field. You can select any number of fields where you would like the system to recognize the user decision to split documents because of a change in the field, such as the PO Number field, or VENDOR field. For our example, we are setting this parameter for the Invoice Number, so when the invoice number changes, the system can capture that the user has separated the document.
1 Like

I am OOO now, but this has to recognize the invoice number to be usefully it has to split there , so I would be surprised there was not a way to split on unique versus every

What I’m finding (potentially) is that it is splitting our invoices when a new invoice field is found.

For example, page 1 of our invoice looks like this:

Page 2 looks like this:

It is the same invoice number, but it seems that because the field is in a different location, it is splitting the document…

Could you elaborate on this? Would this be a change to the DFD?

I would make sure you have Invoice # as a keyword, it should link on the key word to find the invoicenum field.

If the Invoicenum is in another location on the document, that may be needed to train the document differently.