Starting Your OCR Search

shutterstock_298905803_resized.jpg

“Oh, I’m just starting to research OCR” are common words I hear as a representative of a company that specializes in the software. There are some essential items to know when starting your search that will allow you to get the best answers from a solutions specialist (or software salesperson):

1. Know the concept of OCR. OCR stands for “Optical Character Recognition”, but it’s essentially software that grabs text off documents.

2. Know the type of document you want to process:

a. A “Fixed” (also known as “Structured”) document is one where you expect to extract fields that always are in the same location (surveys, questionnaires, etc).
b. A “Semi-Structured” document is one where the fields are somewhere on the document, but not always in a fixed place (invoices, mortgage documents, etc).

3. Know the volume of pages you expect to process. Note that we want pages, not documents. For example, a document may have multiple pages. Most OCR software is sold on the volume of Pages per Year (PPY) so you may need to annualize your estimate.

4. Know how many users will be using the software. This is important because it changes the complexity of the installation. To clarify, a 1 user system can have the software installed directly on the user’s machine. These projects typically have shorter implementation times. Multi-user systems will need a server and database to manage the work. These projects typically have a little longer implementation times.

5. Know what you expect to happen with the output. OCR software will provide two things to you when it’s done: (1) the data/fields that it found, and (2) the document that it processed. For smaller implementations, it’s common for us to store the data in Microsoft Excel and store the documents in a computer folder. For larger implementations, it’s common for us to store the data in a database and store the documents in a Document Management System (DMS) such as SharePoint.

6. So now that you can provide the responses to the items above, know that the cost of OCR software differs greatly on the expectations you have on quality, scalability, and ease-of-use. To validate that the software is within your budget, share your responses with the salesperson and allow him to give you a “ballpark” number. This will ensure that both parties are aware of the high-level numbers and neither are wasting time.

We hope that makes you feel more comfortable with your OCR software search. Please contact us if you have any questions on the details we shared with you or if you’re interested in learning more about how OCR can change your world!

Leave a Comment