Well, a general overview is as follows:

We have thousands of old PDF files which are certificates of analysis for chemicals we have sold. I am wanting to import all of them into Laserfiche and identify key fields on them including customer, purchase order, sales order, and most importantly a multi-value field for batch or lot #. The formats are not standard across the documents depending on customer and product.

My thought was I could run them through quickfields and identify these fields and put them on a template for each document and greatly simplify our storage, identification, and trace ability for these documents. Would there be an option to do this with import agent and workflow? 

If the PDF files already have text files with them, then yes.  If the PDF is just images, they you would either have to use quick fields or if you have distributed cluster computing set up then you could schedule an OCR through workflow.  Then go back with another workflow and read the text file.

If you are trying to set up OCR in the pre-class section, you would do this on an imaged file not an electronic file.  If you plan on getting the information off the page through OCR, you would not need the sample page to be an electronic document.  You would want an imaged document.