What sets ABBYY apart from all free-to-use OCR and some paid OCR software includes the dictionary support for better recognition, which covers a wide breadth of languages, the ability to fine-tune recognition using table, text and image recognition areas, and generally reliable identification of headers, footers, footnotes, etc.
The software is smart about flagging up uncertain characters, but the accuracy is surprisingly high by standard, even on low-quality pdfs.
As a bonus, buying FineReader also allows you to use ABBYY as a professional pdf tool, containing most of the features that you would pay for a full version of Acrobat to get. In that sense, it's effectively two products in one, and the pdf redaction, locking, and electronic signature functions have proven incredibly useful. Review collected by and hosted on G2.com.
Like all OCR software, ABBYY is far from perfect. In future versions, I would like to see smarter recognition of positioned text. As an example, when left aligned and centre aligned text appear on the same horizontal line, the programme should be smart enough to convert this as no-outline table cells or similar rather than resorting to fiddly text boxes.
In general, it would benefit from more "common sense" in the formatting. Rather than copying exactly what it sees to the millimetre, it would be better to look at it holistically and infer how the original document was formatted. Review collected by and hosted on G2.com.
