Monday, March 28, 2011

The intelligent archive — going beyond intelligent capture

By Wendi Klein, director, marketing & communications, North America, for A2iA

Regardless of your industry of focus, you are sure to have heard the word archive. But what does archive really mean? And does it mean the same thing to you as it does to the person in the next office? Is an archive just a repository where documents are stored, never to be found again? Or is an archive something that can actually provide a benefit, or even better, a measurable ROI?

Because of the high demand for timely record retrieval, organizations both large and small need to look to a content management system that will enable them to securely and accurately store their records with as much information as possible so that the documents can be recalled quickly, creating an intelligent archive. This involves utilizing technology that can locate and recognize varying writing styles or mixed document-types and layouts so that once digital, the information can be searched, with keywords or phrases identified for fast retrieval.

Considering that documents are being imaged, how do you then make the documents intelligent —meaning searchable and reportable — and the archive a beneficial tool for the organization? Data capture and routing is critical, although not an easy feat for most recognition technologies and something that requires advanced capabilities — beyond simple rules-based classification or common OCR or ICR.

Great strides have been made in the ability to automatically locate, extract, search, and index data from electronic documents including those of an unstructured format. Technology now allows digitized documents to be analyzed and indexed on a holistic or transaction level in relation to one another, as well as by their geometric layout and content characteristics. The documents can then be searched for pre-defined elements or keywords and, in addition to being archived, the results may be incorporated into pre-existing discovery, redaction, declassification, or document management systems thus providing unparalleled access to the information.

Users maintain privacy and adhere to compliance regulations, as complex and handwritten documents are no longer a bottleneck requiring manual processing. Advanced technologies make a significant difference in the efficiency of the tasks that were previously performed by hand, by allowing the processes to become automated from initial capture through to archive.

Many still believe that unstructured or complex documents can only be keyed and, given the demands of today’s market, any automated solution must be at least as error-free as the manual processes it replaces. However, accuracy is equally as important as successful discovery, due diligence, compliance, and declassification — features found in today’s more advanced recognition and classification engines.

Additionally, those looking to implement such an archiving solution must also examine the definition of and metrics around success. The question should not only be, “What is the read rate?” but also, “How much can be automated, how much time can be saved, how much manual labor can be eliminated, and how robust and comprehensive of a database can be built for search?” The ROI produced for organizations adopting this technology is not only seen in terms of a savings on their bottom line, but also in terms of the time saved through newly realized efficiencies. And once scanned, complex data that was automatically located and extracted can be entered into the IT system and more quickly distributed to those that need it, as well as indexed for archive and retrieval based on complex queries within the newly built database. This increase in accessible and searchable information from a central repository not only speeds knowledge distribution, but it elevates the organization’s global intelligence.

With new regulations and the continued movement towards the paperless office, organizations must consider more than just how to get their documents into electronic format, but what they will do once these documents are scanned. By utilizing the right tools for capturing all data and indexing all documents, an archive can easily make the transition to intelligent archive.

What do you think?

No comments: