Enterprise Content Management: The StrategyEnterprise Content Management: The Strategy

Two important and thought-provoking ideas were recently put forward by Content Management Professionals, the user community launched at http://www.cmprofessionals.org/ late last year. The two credos are that "content management is a verb, not a noun," and that "ECM is a strategy, not a product."

These views are important because they encourage user organizations to stay focused on what they need to do and how they need to do it before they decide what they need to buy. They're also sparking some fresh thinking in what analyst Nat Palmer of Delphi Group recently called "a white hot market that has no direction."

Palmer was alluding to the fact that although studies are pointing to keen interest in content management (CM) these days, there is a paucity of vendor vision beyond "buy or get bought." Once you have all the elements of enterprise content management (ECM) — document management, workflow, report management, Web content management, digital asset management, etc. — under one big tent, what then?

Focusing primarily on Web content management and multichannel publishing, CM Professionals encourages user organizations to base their approach on their content, not on the technology used to manage it. This brand of ECM strategy is all about finding ways to create content once and then reuse it in multiple channels of communication, be it multiple (and global) Web sites, product documentation, marketing materials, call center scripts or self-service knowledge bases. This saves money — sometimes millions — by eliminating redundant content development and translation as well as inconsistencies and lack of coordination from channel to channel.

If there is a technology-focused ECM strategy, it's about establishing a single, consistent infrastructure upon which you can build any CM application. I'm not talking about one-stop-shopping here; I'm talking about building a unified architecture with a single repository (for simplicity and lower cost), consistent GUIs (for lower training costs), unified metadata and search (for easier integration, aggregation and reuse of content), and consistent support for platforms and development languages (for lower deployment and ownership costs).

One could argue that service-oriented architecture (SOA) and the pending JSR 170 Java specification will help vendors and users pull together disparate content management components. But these tools are better at integrating ECM with everything else in the enterprise than they are at papering over internal inconsistencies.

Business users may care less what goes on behind the scenes, but IT knows that a consistent, unified CM infrastructure saves money. I slammed FileNet last week for taking a while to build its own collaboration, records management and e-mail management tools, but the upside for IT is consistency and lower cost. In the bargain, developers and users can combine features for a better application. Web content management or document management apps, for example, can make use of solid workflow and collaboration (or vice versa).

If you survey today's ECM suites, you'll find that the more unified the product, the less likely it is to handle all content management challenges. If a unified product happens to cover all the bases you need, great. But many users still must put together everything they need, accommodating disparate legacy systems in the process.

In the coming years, one twist on the unified architecture strategy might see a separation between content infrastructure and applications. Recognizing that platform vendors IBM, Microsoft and Oracle now have or will soon offer basic services, including check-in/check-out, version control and records management, Dan Ryan, executive vice president at Stellent (www.stellent.com), shared with me last week that his company is preparing for the day — years off, he speculated — when customers might choose one of the biggies for their core repository. Yet these organizations could still turn to specialists for specific horizontal and vertical content management applications. All Stellent and other vendors need to do is design their content apps — Web content management, customer knowledge bases, case management, etc. — to run consistently on top of one of these repositories (or databases with repository services, as the case may be).

I think this idea has legs. In fact, IBM is essentially doing this internally by building JSR 170 interfaces into its many specialized content management products so it can interoperate with WebSphere and DB2 Content Manager. And some content management vendors already rely on Microsoft code for basic library services. Somehow, I don't think I'll be waiting too long for press releases announcing "certified" content apps designed to run atop Oracle, Microsoft and perhaps even IBM infrastructure.

— Doug Henschen, Editor, Managing Content

RESOURCES:
http://www.rockley.com/whitepap.htm
http://www.cmprofessionals.org/
http://www.aiim.org/poster/puzzleposter.html