by C. Roy Payne

Chances are you're processing ever-increasing volumes of documents daily to manage and grow your business. And if you're processing them manually and in hard-copy paper form, chances are this is costing you ever-increasing amounts of money. If you're not converting your hard-copy paper documents into digital images, you're missing out on significant benefits, like e-discovery and disaster preparedness, enterprise-wide document access and storage, less manual paper handling and routing, and so on. Moreover, data capture technologies have come a long way over the past few years, so if you're still manually entering data into databases and applications, whether from paper or from image, you're likely wasting valuable time and money.

Much like manual paper handling, manual data entry can be labor-intensive and costly, affecting the speed and efficiency of document processing, workflows, and ultimately, your business processes. Instead, by implementing advanced data capture technologies to automatically extract information from forms and documents - after they're scanned to image or during the scanning process - you can speed up your processes, service your customers better and gain a competitive edge. You can use optical character recognition (OCR) software to extract data from document images to (a) reduce the problems and costs associated with manual processing and (b) fuel your enterprise and line-of-business applications for indexing, enterprise search and workflows.

This issue of The AccuView focuses on automated data capture and the intelligent extraction of data from forms and documents. We offer various solutions of this nature. From document scanning to automated data capture to electronic workflows, AccuImage provides comprehensive solutions that get you well on your way to peak business performance.

I invite you to contact me to schedule a meeting.

Best regards,

C. Roy Payne
roy.payne@accuimagellc.com

The state of our economy has been headline material for months. Many companies in various industries have seen profits fall. To weather this economic storm, personal finance experts have recommended boosting emergency savings funds and obtaining credit before it is needed. But how is your company planning to remain profitable in the current state of our economy?

We've all heard the terms "do more with less" and "tighten your belts." But what do these terms really mean? According to IT analysts, a typical business reaction in an economic downturn is to cut spending. Many companies have implemented a freeze on capital expenditures and are delaying projects. However, in the case of investments in technology, putting products on hold may do more harm than good. When the economy is struggling and competition is fierce, the infusion of technology into your company may actually give your bottom line a boost. This is especially true for automated document and data capture technologies because they are designed to help you "do more with less" by replacing manual processes.

Consider the typical process. A company must pay employees to manually sort and route all incoming documents. Then, in order for the information from the documents to be usable, it has to be hand-keyed into the company's ECM (enterprise content management) or ERP (enterprise resource planning) system before the document is finally filed. In the case of remittance, a deposit can't be prepared until this manual chain is complete, which can create delays. Furthermore, if errors are made during the manual data entry phase, time is spent retrieving the original paper document from its physical storage location to be reviewed and corrected.

You may be thinking this sounds insignificant, but just one part of this manual process can cost your company dearly. For example, an International Accounts Payables Professionals survey showed the cost to manually process a vendor payment more than tripled in a five-year period. Also, at a recent AIIM International road show, it was mentioned that companies are spending $20 in labor to file a single document, $120 in labor to find a misfiled document, and $220 in labor to reproduce a lost document. Rather than planning for these types of labor costs in your annual operating budget, wouldn't it make sense to replace that reoccurring line item with a one-time investment in automation? In many cases, payback on an automated solution can be achieved in six to nine months, and the operational cost savings will continue.

Automated document and data capture technology has advanced to the point of being able to process nearly any type of business document, including invoices, remittances, checks and insurance notices. With this technology, documents can be scanned with minimal or no presorting required, creating an automatically indexed image of each document for easy electronic retrieval. And, because manual data entry and validation are replaced by automation, employees who previously were required to key in data from paper documents can be reassigned or reduced. The flexibility of today's software means it can easily enforce custom business rules, accommodate ebbs and flows in document volume, and support existing business process workflows. In addition, the captured data and document images can be quickly transferred to nearly any existing ECM or ERP system. The combination of productivity gains and solution flexibility brings value to the bottom line - hence the ability to "do more with less."

Although it may be tempting to wait for the current economy to turn around before adopting new technology, you need to understand that if you wait to see what everyone else is doing, you may be left behind. There's an old adage - "If you see the bandwagon, it is probably too late." While your company is holding back, your competitors may be implementing money-saving technologies to boost their productivity. If you wait for the economy to rebound before you invest, it may be too late for your company to bounce back.

[Source: Integrated Solutions]

Introduction

For most organizations, information has a split personality. It can be the foundation for competitive differentiation - from faster processing time and reduced operating costs to quicker access to information and ensured compliance. Or by sheer volume and complexity alone, it can thwart productivity, waste time and resources, and strain the IT infrastructure that supports it.

The key to utilizing information successfully - rather than being overwhelmed by it - is the ability to efficiently capture and manage large volumes of information from disparate sources. Business information arrives in many forms and must be transformed into intelligent content that can feed enterprise and line-of-business applications, such as enterprise content management, enterprise resource planning, customer relationship management and other information systems.

Intelligent capture enables organizations to cope with enormous volumes and varieties of documents and accelerate information time to value. Intelligent capture includes tools for automatic classification of all document types, extraction and perfection of data for use by enterprise and departmental systems, and tracking that enables document audit and control.

Not a Silo Solution

Document capture can no longer be thought of as a technology running in a silo location; rather, it is an integral part of a business process that can be leveraged at any place during a business process. Today, intelligent capture has moved beyond its status as a standalone application to become a service that can be leveraged by many business areas within an organization.

Intelligent capture goes beyond traditional centralized batch capture and supports both centralized and distributed branch office scanning, enabling organizations to leverage the benefits that come with moving capture out to the point of entry for most documents. Advanced document identification and development tools are provided as part of an intelligent capture solution, enabling you to quickly integrate capture with critical business systems and processes.

The Five Capabilities of Intelligent Capture

1. Capture

Although capture has become synonymous with document scanning and imaging, in the intelligent capture ecosystem, the term has a much broader meaning. Its basic function includes the conversion of hard-copy documents into digital form through scanning, since many critical business processes continue to rely on paper documents. But capture also encompasses the collection and conversion of any structured, unstructured and semi-structured document into electronic data, regardless of source - scanner, fax, file system, e-mail or electronic file.

2. Classify

In today's enterprise, documents of all types are captured across the organization. Automated classification technology distinguishes between multiple document types - such as invoices, order forms, claims and loan applications - without manual separation, separator sheets or barcodes to identify documents. It employs a combination of text-based and image-based analysis to identify document types. Text-based analysis recognizes documents using keywords, phrase and word combinations, while image-based analysis identifies documents by format. Advanced solutions can also differentiate between similar documents using a logo or comparable graphic indicator, and even recognize handwritten documents from machine-typed documents.

Auto classification enables intelligent routing of documents as part of business process workflows and significantly reduces the manual preparation that organizations have traditionally employed large staff to handle. Automatic classification also enables the assembly of complex documents such as loan origination packages, minimizing user intervention.

3. Extract

Once documents are electronically captured and classified, their data is extracted using optical character recognition (OCR) engines for machine, handprint and optical mark recognition (OMR) such as checkboxes and bubble fill-ins. Data extraction can follow business rules and reference libraries of keywords, word associations and syntax structure. Various recognition engines can be utilized, allowing engines to be leveraged based on their strengths with various types of data. The technology can also automatically detect tables in documents and extract data from rows and columns - analyzing line, column and page breaks to manage items appearing on consecutive pages. Further refinements of extraction include purchase order and general ledger account matching, and support for complex forms like health care insurance claim forms. When necessary, data can be extracted manually. For example, in a key-from-image operation, an operator inputs the data from an image if image quality makes automated extraction difficult.

4. Validate

Data validation ensures data was extracted correctly. There are two types of validation: automated and manual. Automated validation involves integration with another data source, usually a database or enterprise application such as CRM or ERP. The extracted data, such as a list of account numbers, is compared to a known set of data to ensure accuracy. The same process could be performed with multiple data fields, such as information on a health care claim form. Manual validation includes character correction, which allows operators to review and correct data that has been extracted with a low level of recognition confidence.

Intelligent capture solutions vary in their ability to simplify the validation process. For example, automated validation can be accelerated by performing database lookups or through other business rules. Operators are notified when a rule has been breached and data requires manual validation. Additionally, read-only fields in the extracted data can guide operators to minimize rework during manual validation, skipping over fields that do not require review. Moreover, keyboard entry can be eased with hotkeys that are customized and tuned to specific data sets.

5. Deliver

Data and document exports are the way intelligent capture makes information available to other systems. On the simplest level, data can be exported to a file system in a variety of data and document formats, including XML and delimited. Scanned images can be stored as PDF, JPEG, TIF and other formats. Another export alternative is to leverage ODBC, which makes data compatible with any ODBC-compliant system - Microsoft SQL Server, Oracle, IBM DB2, etc.

A third option, enterprise system exports, delivers the greatest range of flexibility, connecting capture to business systems, content management systems and processes. These exporters enable an intelligent capture system to become an extension of other enterprise applications - transferring data as well as images, triggering workflows and participating directly in business processes. An intelligent capture system without this level of integration is not suitable for most large organizations, which often rely on several mission-critical information systems.

Deploying an Intelligent Capture Solution

Whether you need an intelligent capture solution to meet general information capture needs or require a solution that is proven to handle high volumes of critical documents like loan documents, insurance claims and new account applications, make sure the solution is complete in its offering and addresses all key aspects - distributed processing, intelligent document recognition and process integration. Design and development should follow a best practices methodology that includes measurable goals and objectives, and quality assurance testing.

[Source: EMC]

Q: What is data capture?

A: Data capture is the method of acquiring data off of paper forms and documents. You can process forms and documents using manual data entry, by keying data directly off the paper document into your back-end system such as an enterprise resource planning system, or with an automated data capture solution (using software that captures the data automatically) for transfer into any business application.

Q: How do I know if my organization needs an automated data capture solution?

A: If your company needs to capture data off the hundreds or even thousands of incoming paper forms and documents it receives each day, week or month; if you rely upon manual data entry to capture this data into your systems; if your company devotes significant personnel resources to perform this manual data entry ... you need to automate. Processing documents manually can incur thousands of dollars in hidden costs for your organization. Offsite document storage and retrieval, lost revenue opportunities due to document processing backlogs, and human error significantly increase your price to process documents this way. Many companies save hundreds of thousands of dollars each year by switching to an automated data capture solution.

Q: What are the differences among structured, semi-structured and unstructured documents?

A: Data capture technology has traditionally handled structured forms, such as order forms, time sheets and surveys. These items have a uniform format where data is found in precisely the same location on each document. The customer name, the date and the item number are always found in the same spot on each order form, for example.

Semi-structured refers to documents such as invoices that every company receives daily from multiple vendors. Semi-structured forms have the same type of information on each of them (invoice number, invoice date, total due, etc.) but depending on the vendor, the location of that information could be anywhere on the page. For example, one invoice might have the invoice number in the top right-hand corner of the page while another invoice has the invoice number placed in the middle of the page. Software products that perform semi-structured processing must be able to locate and extract the desired data, regardless of where it is located on the page.

By comparison, unstructured forms are a complete free-for-all. There is no consistent format, look or requirement for unstructured forms. Letters, e-mail messages and direct mail are good examples of unstructured business documents.

Q. Are structured and unstructured documents processed differently?

A: Yes. Since data placement in structured documents is consistent, a template is created which assigns zones where the data on the document will be every time, and you then define what types of data will be contained within these zones. The location of data on unstructured documents, such as invoices and Explanation of Benefits forms, is inconsistent and requires something more flexible than a template-based solution. Take a purchase order number, for example. Software searches the document for "P.O. No.," "P.O. #," and other variations of the term. Once it locates and captures that data on the document, it remembers the location each subsequent time that particular vendor's invoice is processed, thereby accelerating processing time.

More questions?

Great! We look forward to answering them! Please call AccuImage at 615.242.7226 to discuss your data capture requirements and get started on the path to automation today.

[Source: AnyDoc Software]