What is Document Classification Software?

Document Classification Software is able to identify and sort documents into groups using intelligent page layout and keyword analysis. The more complex applications have interactive training capabilities that learn to recognize new documents automatically. Less expensive applications are also available that require more document-specific configuration but work well for many classification tasks.

Who can benefit from document classification software?

Companies that need to digitize or capture data from many different types of documents that do not come pre-sorted should invest in document classification technology.

Enterprise Mailroom Automation is the classic example for a classification solution. The system is trained to identify and label all the documents commonly received throughout the organization and route them to the correct person or workflow electronically. These types of solutions are best suited to mid to large sized companies with very high volumes of documents.

Smaller businesses and departments can use less expensive classification solutions to perform a variety of useful tasks. The key question is whether you can use keywords or templates to identify and group your documents. If you can define a unique list of key words or phrases that identify a document there are very cost-effective solutions available. If you have a limited set of documents to identify and can provide a sample of each, then template-based solutions are available that are more robust but still affordable.

How much do Document Classification systems cost?

The total cost of a Document Classification solution includes several items:

• Cost of the software
• Time to install and configure the software
• Configure classification rules and train classification engine
• Data Capture Workflows configured for automatic filing or data entry
• User and administrator training
• Labor required to verify classification results
• IT infrastructure and maintenance costs

The major factor in software cost is the classification technology that is employed. Without configuration, keyword-based classification software can be purchased for around $1,000. Template-based solutions can be found for under $5,000. Enterprise solutions start around $15,000. However depending on the volume and types of documents involved, and especially the number of Data Capture Workflows that are integrated into the process, the cost of configuring and testing will be the biggest influence on total cost.

If you have an IT staff that is familiar with document scanning and OCR applications, it is possible to do most of the configuration and maintenance in-house. If not then it is highly recommended that you use our Consulting Services to guide you through the setup process.

Contact Us to get a professional analysis of your project requirements and a full time and cost estimate.

What is the typical Document Classification workflow?

Using the Enterprise Digital Mailroom to illustrate the most comprehensive type of implementation, the following steps are involved in the solution workflow:

• Mail is opened and prepped for scanning (unfolded, staples removed, junk discarded, etc.)
• Paper documents are scanned on a high-speed document scanner
• Electronic documents can be input via email or hotfolder
• Classification engine identifies the documents
• Unknown or uncertain documents are verified by a user, further training the system
• Documents are routed to appropriate data capture workflow
• OCR, ICR, checkboxes and barcodes are used to index or enter data from documents
• Uncertain data fields are presented to users for manual review and correction
• Once all errors are corrected, data and documents are exported
• Data is sent to relevant databases, and/or used to file documents automatically in your file system, SharePoint, or document management system
• Documents can then enter electronic workflows for routing to email recipients and performing any necessary reviews or approval processes

There are many steps involved in the process, including some manual review, but the end result is that the vast majority of paper in the organization is digitized, and all associated data entry and filing tasks are automated.

Simpler solutions combine the classification and data extraction steps, and don't interactively train the classification engine.

How do I find out more?

• Overview of Document Classification software products
• Forms Processing Software product catalog
• Compare Features for Forms Processing & Data Capture Systems
• View our FlexiCapture for Invoices Case Study
• Contact us to get a professional analysis and estimate for your project

Some common Document Classification solutions we have implented are:

Enterprise Digital Mailrooms: complete digitization of all documents in the organization, as described above.

Mortgage & Loan Processing: identify the hundreds of forms in a mortgage file to ensure all required documents are present, and to sort and export in pre-defined "stack orders" required by the banks. This solution is available as LoanStacker.

Tax Preparation: identify the forms and other documents in a scanned tax return, creating a PDF with the sorted return and bookmarks for each form for quick reference.

Human Resources: use keywords to automatically identify qualified candidates among thousands of resumes.

Legal Disclosure: automatically send new case documents to relevant parties as required by law.

Mixed Document Capture: avoid manual sorting of paper when scanning mixed batches of documents.

IRISXtract is an Enterprise solution designed to manage every aspect of the document capture, classification, data extraction and workflow routing process.

IRIS PowerScan includes a deskop version of the IRIS Document Classification engine for a much lower price.

ABBYY FlexiCapture is a template-based solution that lets you define a variety of content elemets to identify documents. It also has complex data capture capabilities to automate data entry tasks associated with these documents.

SimpleIndex is able to identify documents using unique keyword combinations, at a fraction of the price of other solutions. It has a unique, open processing workflow that lets you peform a variety of scanning, classification, data capture and routing tasks in a single automated workflow, but without the need for expensive server infrastructure.

Compare Features for Forms Processing & Data Capture Systems

	How do I keep a running count of the number of pages scanned?
	Can you use OCR to read data from documents if it is not in the same place on every page?
	Can I scan a page with tables or form fields and have the output saved to a Microsoft Excel spreadsheet?
	What is a product that will allow me to create forms and scan completed forms (handwritten) into a database or spreadsheet?
	Assuming a load of 500 documents per day, which Forms Processing software would you recommend in regards to value, reliability, and ease of use with a mid-volume scanner?
	We want to automatically scan different forms, recognize what form it is, and file it for later retrieval. What software do we need?
	Does ScanStore provide outsourced document scanning, OCR or forms processing services?
	Can I scan a completed form into my computer and change the information, written text, or layout of the scanned document?
	I am looking for a software package that will allow me to scan a PDF document and fill in the blanks for electronic invoicing. It appears that your software will allow me to do this. However, I am also looking to have our invoices electronic requiring customers to sign the invoices using an iPad. Will your software package allow a free hand signature as well?
	I would like to scan a document, then fill or type in the blanks. Will OCR software allow me to do this?
	Please let me know if the price stated in the Web site in US$ dollars?
	Is it safe to use my credit card on ScanStore.com? How do you protect my information?
	What is your return policy? If the product has a problem on delivery how do you handle that?
	Can scanner capture the image of document with shaded part of the document and still recognize information from that section with OCR, ICR or Barcode?

All FAQ Topics