Those with doubts about the definition of content management can rest assured that experts agree on one thing: content management is poorly defined. Sometimes it is Web-related, at other times about data. Debates still ensue about whether true content is frontend or backend.
Despite such misgivings, enterprises are finally embracing content management at the front end. Intriguingly, some of the more compelling offerings come from technologies once regarded at best as marginal, or worse, as novelties.
"It's been so ill defined and created so much confusion out there. It's so fragmented at the moment that content management means something different to every one. You need to have an enterprise view of content," says IBM DB2 group information mangement specialist Ursula Paddon, who has just completed a market scoping study of the content management intentions of 203 Australian call centres with independent research house ACA Research. While IBM will not release the whole document [media content management], Paddon says the results are a mixed bag.
"It's not rocket science, but once you look under the covers what's stopping [adoption of content management] is poorly integrated IT systems and cost factors," Paddon says, cautiously noting that heavily siloed architectures are putting some backs to the wall.
"Call centres are cost centres and the key pressure is bringing the cost down. But there is only so much cost you can cut out of an operation and that has been done. You can't reduce cost any further; there are certain key ratios you just cannot go under. The question now is how to improve performance of individual agents," she says.
While digitised content management can deliver savings of around $750,000 a year to the average 100-seat call centre, Paddon says customer-facing business processes still rely heavily on paper forms and documents. These can include everything from Medicare forms, insurance claim applications to credit applications contracts and so on.
"In most cases this is still in paper-based format. There is simply a lot of documentation that is still not yet in an electronic format. So the integration task, [which is] difficult enough for data warehouses…when you put on a layer of unstructured data, it's no easy feat" Paddon says.
All of this is great news for content front-end players like Adobe and Scansoft. While Adobe's PDF technology has achieved similar desktop ubiquity to Microsoft, it is only in the last year that Adobe has targeted the business process content arena through the acquisition of server-based forms application vendor Accelio (best known for Jet Forms).
"What we have done is adapt PDF technology around that so that now on the current version there is XML incorporated - so you can extract the content from a PDF and send that to a database or content management system," says Adobe Australia's managing director Craig Tegal.
Known as Document Server for Reader Extension (DSRE), the offering allows creators of PDF files to add functionality that self activates within a file accessed via Acrobat Reader, through Java scripting.
Such functionality includes the ability to save, sign and secure PDF forms that can then be completed by applicants at their leisure. Data can then be automatically extracted from the forms for backend systems by way of XML, which has also recently been incorporated into PDF.
"It offers the ability to take paper-based systems and forms processes and turn them into electronic documents - and be able to manage the process of approval for those documents," says Tegal, adding that governments view the idea of citizens having to buy software to fill in forms with great disdain.
Tegal says a customer-facing federal government department will go live with the interactive PDF offering to customers in the new year, although he not yet at liberty to name it. Westpac is also known to be evaluating the system in an effort to streamline interactions with customers.
Another tool in the business process paperchase finally showing dividends is optical character recognition. For the last decade many large organisations have salivated at the prospect of being able to digitise and mine the billions of paper records archived away.
After a decade on the fringes of success, the acquisition of OCR (optical character recognition) vendor Caere [check SP] by recognition vendor Scansoft confirmed that both voice and optical recognition technologies were coming of age. Ironically, the Omnipage OCR software now under the control of Scansoft has been around for more than a decade, but has only recently found success in the enterprise when bundled with other suitably graded solutions such as voice recognition.
"We are now capturing the layout and that is a dramatic change in what happens when you scan and convert. Images and columns can now be converted over and re-established in [an electronic] document," says Scansoft's Australian managing director Bob Anderson.
The result has been forms clients like the US Inland Revenue Service and Bank of New York. Such forays into the enterprise forms digitisation space is significant because both are clearly now taking such content recognition technologies seriously - and porting them back to solutions such as business intelligence and data mining.
Meanwhile, Scansoft is also pushing the Adobe envelope with file conversion applications such as PDF Converter that can take composite files and port them directly into enterprise data environments such as XML or into Microsoft Office applications. Laughing off suggestions that such technologies are reverse engineering, Anderson is keen to reiterate that both vendors are happy to share the same space.
"The technology recognises the [Adobe] security features, that's integral," says Anderson.
After 10 years on the outer, frontend conversion technologies appear to have found their a legitimate place at the enterprise table - however unheralded or confused it may be.