Last month, we discussed the main functions of a document management system, with regard to documents that are internally generated and, therefore, already available electronically.
This month, we want to concentrate on external documents.
External document management
External documents can enter your company in a number of ways.
- Courier
- Fax
- Hand delivery
The documents received via e-mail are the easiest to manage. They are already available in an electronic format and can be imported into the archiving system and processed from there.
Fax documents are as easy, as well, if the implemented solution for fax allows an incoming fax to be routed to an e-mail address.
That leaves us with the traditional paper documents. They come in many shapes and forms, and they are the most challenging sources of information for a document management system. But, in many cases, they are also some of the most important documents, because they may contain signatures or other handwritten information that might become critical in case of disputes or questions.
All paper documents need to be processed by a scanning device. There are a wide variety of scanning devices on the market. You can buy a simple scanner for less than $100 or you can spend thousands of dollars for sophisticated, duplex, high-resolution scanning devices that can process more than a page per second.
Although, naturally, there is a big difference in speed and quality between an inexpensive scanner and a high-end device, they basically perform the same function: They read a piece of paper and produce an electronic copy of it, which is made available as a file (typically in .tif format) for further processing.
At this stage, the former paper document can be treated the exact same way as a document that was received via e-mail.
The result of the scanning process is a photographic image of the document. This is of high-enough quality to be printed, reviewed or sent via e-mail or fax; however, the data contained within the document is not accessible though the stored, photographic image.
To access the data, we have to re-engineer the document. The technique to do that is called Optical Character Recognition (OCR) for printed or typed content, or Intelligent Character Recognition (ICR) to read hand-printed numbers and letters. As with scanning devices, there is a wide range of OCR/ICR software on the market, ranging from packages for a few dollars to products that cost thousands of dollars.
The largest difference between the inexpensive solutions and the high-end solutions is the accuracy of the data recognition. Where high-end packages will be able to recognize close to 100 percent of the data from a printed image in good quality, you can’t expect more than 80 percent to 90 percent of handwriting recognition, even with quite sophisticated solutions.
In any case, to ensure that you use good data, a quality-control and correction mechanism has to be introduced to your scanning/recognition procedures.
Some packages add intelligence to the OCR process to recognize certain forms, like recurring invoices, purchase orders, etc., and, by doing this, allow you to automate the data-entry process beyond the document management system and make the recognized data available to be processed in the company’s ERP package.
As much as OCR/ICR technology has evolved over the years, it is still a relatively unreliable method to gather data. You should communicate with your suppliers and other external resources to exchange your information electronically to avoid possible errors.
Alex Reiss is president of Resolutions (www.resolutions.ws), a provider of state-of-the-art document management solutions. Reach him at [email protected] or (678) 714-3400.
