First class classification
Getting automated mail room technology wrong is a recipe for disaster.
Companies are often under the illusion that deploying a mailroom scanning solution will improve their business efficiency. But that's not the case. Simply digitising documents is only pushing the paper images around, and doesn't actually remove the real obstruction. If the right technology and controls are not in place, the chances of losing documents can actually be increased.
It's what happens after the scanning phase that counts. This critical second stage of processing - where recognition and classification takes place before any optical character recognition (OCR) is attempted - is the one that has the largest beneficial effect on mailroom efficiency.
So how does one deploy this key step in mailroom automation? And how can the relevant information be extracted from the paper documents as efficiently as possible to unblock the corporate workflow?
Full marks for intelligence
The critical step is the quick classification of all scanned documents, followed by the application of auto-indexing techniques via OCR or ICR (intelligent character recognition) technologies. OCR/ICR is relatively expensive technology to deploy but, in practice, it's only necessary to process on average 1.5 pages of each document in order to correctly classify and index it.
Approximately 80 per cent of the information used by an organisation is in the form of unstructured paper documents and, until recently, automatic processing was limited to documents with a fixed structure.
For organisations that scan and capture their documents, appropriate handling used to involve spending considerable time and effort sorting these documents prior to scanning, adding document separation sheets and manually classifying the document prior to data extraction - meaning a considerable amount of skilled sorting work both before and after scanning.
Auto-classification and recognition techniques further automate document handling by removing the need for intensive, manual post room sorting and document preparation tasks - all of which helps to take a significant bite out of overall processing costs.
This type of intelligent automation software (such as Indicius and Xtrata from Kofax), provides advanced document classification, separation and extraction capabilities. Using OCR it can automatically extract machine printed or hand-printed text and printed data from scanned document images, removing the need for costly and time-consuming manual keying.
The extracted data adds to the intelligence already gained about the document, aiding separation and process routing of the document to a workflow queue, facilitating the indexing of the document for storage in a document management system or sending it direct to a business process. With recipient detail or an identifying case or account number extracted, the appropriate documents can be automatically routed to the right process destination: purchase orders to the sales and manufacturing departments, invoices to the finance invoice processing system.
At the end of the scanning process, images and the extracted data are usually released to a document management system, case management system or image repository to ensure security, safe archiving and auditability. This is vital to avoid the loss of documents. If the document management system links to core business systems such as enterprise resource planning or customer relationship management, this further automates and enhances processes.
So much for the theory: how does this work in practice? Let's take a look at a real-life case study.
Countrywide Property Lawyers is one company that is benefiting from this classification technology. The legal firm deployed an advanced data capture and document processing solution to automate mailroom activity and the handling of the 25,000 legal documents received daily.
The solution automatically allocates 85 per cent of incoming post, using Kofax capture modules, to the correct legal case. It also classifies 90 per cent of post for staff, automatically identifying letters from lawyers, draft contracts and mortgage offers, helping to speed up business processes and ensure legal teams get essential legal documents by 10:30 on the morning of receipt.
The automated mailroom handles all business-related printed and hand-written documents, extracting key data and sending it along with an image of the mail to the case management system, which automatically routes the mail to the correct legal team for action.
Documents that cannot be auto-referenced are then sent as an image file to India, where they are eyeballed and manually categorised inside the Visual Files Case Management system and sent back to the UK to the relevant legal teams. As a result, only five per cent of all incoming mail requires manual intervention in the UK - this has led to a 90 per cent reduction in administration.
In conclusion, quick, automated classification underpins a successful mailroom automation solution for three good reasons.
First, it's simply good business sense: it cuts transaction costs across the organisation by sorting the wheat from the chaff. Second, it improves relationships with customers, by accelerating transactions and ensuring records are kept updated. Finally, it also addresses expanding information management legislation, which specifies the need for companies to capture, track and control information - especially financial information - as soon as it touches them.
There's never been a better reason to start sorting the post more efficiently.