This is a common challenge, as the PDF software used by copiers often creates the PDF so that text isn't actually "regular" text, but is instead a picture that happens to look, to you and I, like text. But software like Monarch that searches the file for text won't find any.
I did post a [url="http://mails.datawatch.com/cgi-bin/ultimatebb.cgi?ubb=get_topic;f=1;t=001155;p=1#000000"]possible solution[/url] to this type of challenge awhile back. Depending upon the software you have available to you, this type of approach may be useful.
Unfortunately OCR software not a common tool in many offices.