Friday, September 21, 2012

Gartner, Forrester and Doculabs have all been following the Integrated Document Archive and Retrieval System (IDARS) marketplace since all the way back in the early 90’s.  The concept is not new: store high volumes of internally generated content in a highly efficient system with a database for metadata, and separate storage for documents themselves.  With security, allow internal users and external customers access to the database, and the documents that the database points to. 

The concept has been a smashing success.  Virtually all mid-to large sized enterprises have deployed some version of IDARs to reduce printing costs, make content available to CSR’s so that they can answer customers’ questions on their statements or bills in real-time, and vendors have built solutions that have scaled, integrated them into total content management infrastructures with federated search across IDARs, and other content repositories. And we’ve all made or saved a few bucks in the process.

So, after 30 years of solid but rudimentary success, you’d think there would be nothing further to do here.  These systems are now embedded within the enterprise, the vendors have built mini-empires based upon the results they’ve delivered, and it’s a pretty static landscape in terms of new developments.  So, it’s surprising that a division of Actuate, squarely in the business analytics space, has decided after all this time to throw their hat in the Repository ring as well:

There are two compelling reasons why customers want more than the traditional IDARs: User Interactivity and Cost Allocation.  Let’s first discuss why users are bored with their current IDARS:

 Internally generated content like reports, customer statements, insurance claims documents and complex financial reports are not an island unto themselves.  This content fits into a framework of information that organizations are increasingly mining to make decisions, predict behaviors, and satisfy broad regulatory requirements.  I recently visited a top five P&C insurer, and the news is that first tier organizations are no longer content with content as usual.  They are thinking about the requirement to store underwriting documents, and all the process-related decisions that generated the documents, for example.  They need to generate a view into the rules that generated the decision to underwrite a policy with a specific risk calculation, so they can evaluate that decision against actual claims data mined from data warehouses and claims reports.  This is not your granfather’s IDARS or ECM system anymore.  We are talking about merging, linking structured and unstructured content and being able to present not only views of documents, but more interactive views of rules engines, documents and compare predictions with data actuals.

Cost Allocation: Traditional IDARS systems are pretty vanilla when it comes to how they report on who uses the system, what percentage of the content is being retrieved from, say, the Variable Annuities LOB, vs. the Protection LOB.   You get a pretty standard set of admin screens and you basically have to build your own reporting based on system logs.  Because ECM is truly an enterprise application these days, set up centrally to manage content from all business units, the need to do this reporting is key from a “who pays, and how much” standpoint.  What if you could generate dynamic reports that allow you to generate dynamic reports that show usage be department in terms of total bytes stored, total numbers of retrievals, which departments added the most new reports, and do all of this with a simple, no-programming interface?  The answer is that you’d be able to better allocate costs, but you’d be able to better predict what your costs will be for new applications, because you can easily view what’s happened in the past.  And, there is more going on here than simply the ability to generate beautiful, 3D graphs, and interact with the reports, we are also talking about being able to make associations between the granular data you store, how it is indexed, and how it ties into records retention.  It is not good enough to store and records manage “corporate financial reports.” These reports relate specifically to individual corporate entities, and people, and if you have a system that can see those associations, and allow you to manage at that level, you get a much more elegant content delivery platform, and you will attract user communities that know they will get value based upon actual usage.

Figure 1: dynamically generated reporting about what’s in the repository

And this brings us to why an analytics company announced a new IDARS system.  Today’s ECM systems are primarily focused on systems of record.  Store unstructured content, and serve it up to users who request it.  But, as the insurance company told us, those are yesterday’s requirements.  What’s needed now, is a system of engagement that can store, link and dynamically present information derived from a variety of structured and unstructured content, and to do so over web, mobile and touch-tablet device channels.  The Next Gen Repository discreetly manages content by the audience that views it, not just by the name of the content.  Please stay tuned as we uncover more trends on this topic.  One thing’s for sure: yesterday’s methods will only produce yesterday’s results!