As new technologies like web services and methodologies like service-orientated architectures (SOA) converge with the desire for unified communications, the need for a more coherent and comprehensive information strategy is only going to increase.
As far back as 2004, Gartner began researching and writing about this strategy and started calling it enterprise information management (EIM). The idea itself is not new. For years, business people have wanted a single view of the customer, for example, that displayed real-time information that was cross-referenced across all the customer’s accounts regardless of geography, and, most importantly, a view that displayed quality, reconciled data.
Today, that desire still exists since the technologies of the last decade, while improving many aspects of the business world, have not done much to bring together (and reconcile) all the disparate silos of corporate and customer data.
Most business people still need to cross check information to ascertain accuracy and, even then, there are often discrepancies created because of things like semantics, time-delays, incorrect data entry, and multiple, unsynchronized data repositories for the same information.
This leads to a lot of wasted effort and time lost, said David Newman, a Research VP at Gartner.
“It doesn’t make sense to have 10 different versions of a product hierarchy, or different definitions of customer or different definition of sales,” said Newman. “I can’t tell you how many meetings—before I was an analyst—on the operational side; when you go into these CXO meetings and they’re not arguing about what to do with the information, they’re arguing about whose data is right. That’s not a very effective use of those people’s skills.”
EIM is an attempt, like program management for IT projects, to put sound governance around the way information is viewed, used, collected, disseminated, accounted for, etc. and applying those policies corporate-wide.
And, as applications become more loosely coupled through SOA initiatives, the information those application consume will also become disassociated making matters much, much worse. No longer will one application rely on one data store to do what it does. Data will come from anywhere—everywhere.
“One of the biggest barriers to information access in the enterprise is the fact that data is often stored in so many different repositories, wrote Microsoft Chairman Bill Gates in his May 17 Executive Email blog.
“This leads to painfully inefficient processes that force information workers to leave one application, logon to another, find a single piece of data and write it on a piece of paper, and then return to their original application, just to complete a simple task like sending an email to a customer. This is a significant drag on productivity.”
Gate’s view of EIM includes a world where “searching” is no longer something you do because an effective EIM will break down disparate data stores and workers will have what they need at their fingertips regardless of the application they are working in.
Enterprise portals were one of the first cracks at this type of convergence but they fell short. Not because they didn’t deliver on what they promised, said Newman, but because they only group applications in one place, not the data that supports them.
For an effective EIM you need to bring together all of the elements of your systems, infrastructure, policies, processes, procedures, people, semantics, etc., etc. It’s a long list and one not easily contemplated let alone implemented.
“If you’re going to do EIM you need to address governance, you need to address the organizational implications, you need to address process, you need to have a common infrastructure, you need to tie it to your business strategy, you need to have metrics to measure the overall effectiveness and performance, said Newman.
“So we see it as not a technology, not a market, but as a series of building blocks, a series of integrated disciplines.”
If you’re wondering at this point why ILM (information lifecycle management) doesn’t already address some of these issues, you’re probably not alone. Like many IT initiatives with new-sounding names, there is confusion between the terms knowledge management (KM), business intelligence (BI), ILM and EIM. Parts of each discipline can overlap but not necessarily. It all depends on the particular problem being addressed.
Anders Lofgren, SVP Product Management for CA’s Storage Management unit, sums it up this way: “One way of looking at is EIM is the front-of-house kind of thing, to use a restaurant analogy, where that’s actually dealing with the direct consumers, the direct users of the data.
“Where ILM is really a back-of-the-house kind of thing where it’s actually dealing with the operations with regards to infrastructure and things of that nature.”
And, even though there is some confusion for now, on the bright side EIM will not necessarily require new tools, but the combination of existing tools—business intelligence, enterprise content management, and enterprise search in particular—to manage data (and metadata) in a way that optimizes its availability and use across the organization.
“What’s required,” wrote Gates, “is a comprehensive approach to enterprise information management that spans information creation, collection and use and helps ensure that organizations can unlock the full value of their investments in both information and people.”
EIM aims to do just that.