Expand OCR capabilities within Form Extraction
Perhaps this already exists - I am not aware of it existing - but while I love the Form Extraction functionality, I find that there are often times where I am working in documents of similar format and presentation but with slightly different dimensions, meaning, Form Extraction works great on a standardized form (e.g., a tax standardized form), but let's say I am importing pay stubs for 100 employees and I want to use Form Extraction to add snips to those stubs and extract data, the moment that one of those pay stubs has an added line (i.e., a one off bonus that was paid in that pay cycle but not all cycles) to the stub, then the Form Extraction is no longer useful. It would be nice if the Form Extraction could be similar to Document Matching in that OCR would recognize when there are slight variations in a presentation/template, but still be able to successfully extract the data I am looking for. Thanks!
Andy G.
shared this idea