While the interface & workflow could certainly be improved dramatically, the bottom line is that this is the BEST OCR out there right now. So far I am getting 99.9% accuracy which is amazing to me. So far I have scanned 2 books this week, one a 120 pager, and the other a 90 page typewritten xerox copy from 1982. The only hiccups were as expected: page tears of the original, sub-par scans, and odd-size text of the first letter of the first word. Even slightly askew scans were easily handled.

The biggest and most important tip I could offer is to first get the best scan possible: for text, scan greyscale, 300 dpi. It should be obvious, but start with the best, cleanest original & scan available. (Garbage in, garbage out... )

The workflow that worked best with its rudimentary interface is to do all my scans first to a folder, then launch FineReader and plow through the folder. You can only save as text or RTF, so choose wisely. Do a test to see which works best for your project.

OCR gets 5 stars (forget PDFPen and Adobe Acrobat, I've suffered with them both), and that is what matters the most. The main rating gets knocked a star for rudimentary GUI that could easily be improved. The $99 price seems a bit much (especially since the Windows version is half the price!) but the accuracy for me makes it worth every penny.