The chief disadvantage of optical character recognition scanning is the potential to introduce errors into a scanned document. No OCR scanning system is infallible, and poor quality documents can create enough errors to require lengthy and time-consuming proofreading. Handwriting and non-Latin fonts are particularly difficult to scan correctly.
Optical scanners can also have trouble with documents that lack significant contrast between characters and the background. Dirty pages, or those printed on colored stock, may confuse a scanner and result in large blocks of unread text. The extra steps necessary to render poor-quality originals suitable for OCR scanning may end up completely offsetting the potential time savings the technology offers.
The best way to ensure OCR scanning is as accurate and quick as possible is to design the original document specifically for optical scanning. For instance, many OCR packages have difficulty differentiating between similar characters, such as the numeral zero and the letter O. OCR-friendly fonts make these two characters appear as different as possible to aid the computer in reading the document.
Handwritten documents require even, clear spacing between letters to ensure proper scanning. This is why forms that use OCR technology often require users to print in capital letters inside a series of boxes. Each separate box can then be scanned individually, preventing the computer from misreading letters that have run together.