I've been using Scribe a LOT lately to automate filling out PDF forms. What I do at first is open the form in question, put in some sample data, and then run a script that finds all the field names and the values with which they're populated, so I can then develop an appropriate script to fill out that form (that's my big tip).

However, what I've learned is -- man, I cannot believe how various forms name fields. My favorite thus far is a patent office form that for a patent application number names the field some boilerplate text that runs to about a paragraph (!). You don't ever see this unless you're doing something like with Scribe, but wow.

yea i kind of gave up on trying to rely on any logical naming of fields in PDFs especially from any 3rd party - they just are abysmal. However I have to agree Scribe is a great tool as you know I use it for looking at email subjects for processing and also use it to change cell names in excel files from table::field to something more legible, critical when you are sending the excel file off to the client or some service.

I was just floored.  This is from a government agency no less.  Tons of misspellings, labeling things as check boxes when they weren't, the aforementioned paragraph as a field name, and one of my favorites -- a lazy developer who just started naming all the fields "undefined1," "undefined2," etc.