If you've been paying attention, you might have picked up a little bit of Spiderman mythology hovering over what we now call "enterprise content management." At the beginning of the story you have the humble but sincere document management. Good at saving money by automating repetitive tasks in the basement of the building. Sort of a Peter Parker vibe: nice enough guy, but kinda nerdy. Definitely skillful and book-smart, but in no way is he going to get MJ's attention. And forget about J. Jonah Jameson. He'd laugh him out of his office in a nanosecond.

Then along comes a spider. The mutation of electronic document management and Web content management, this little monster bites document management one day, and BAM! You've suddenly got an unholy hybrid of puritan work ethic crossed with 21st Century cyber-nautics. The only question: Will he use his power for good or for evil ? For with great power comes great responsibility...

I'm just messin' with you. But it's weirdly true that the maturity of what we now correctly or incorrectly call enterprise content management has followed a mostly accidental evolutionary formula that put business automation in a beaker, added some electronic content requirements and shook vigorously until...

What came out? I set out to learn what, exactly, is just so "enterprise" about enterprise content management...

'The shift has been dramatic in the last three or four years. Customers now commonly think of content management as an enterprise solution, much in the same way they think of email as an enterprise solution, or database applications," says Rich Buchheim, senior director, product management, content products and strategy at Oracle.

Eric Stevens, senior director, industry and solution marketing group, at Hummingbird, follows on: "Examples of enterprise content management, to us, are those companies that value content the same way they value data at an organizational level. It's a tricky definition. Sometimes the CMS is just a filing cabinet-I put Word documents there when save them. But a better example is the company that says: 'How do I use that content to help people do their jobs better?' That's when it's an enterprise application."

"Enterprise means more than crossdisciplines; it also means cross-functional," adds Brian Anderson, vice president of marketing at Siderean Software, adding that ECM transcends the boundaries of mere departments to include information located outside the organization as well. He gives an example: "An aerospace company researching a new project needs information from everyone in their large enterprise who has worked on similar projects (internal). But they also need all the papers and research related to technologies around this kind of project (external)."

Colman Muiphy, senior product manager, Xerox DocuShare, puts the definition into a different context: "Corporate functions (HR, call center, AR/AP) are obviously very different, but they all share the unifying characteristic of involving content that can either represent opportunity or represent risk-opportunity in the sense of protecting corporate investments; risk in the sense of not effectively disposing of content across the enterprise."

Oracle's Buchheim adds that "it enters the realm of 'enterprise' when it no longer solves the productivity (or compliance and risk management) problems of a single function, but facilitates the sharing of information across departments. I can put a CM system in my HR department tomorrow, and it will help the HR department. But when the HR folks and the legal department need to collaborate (and heaven help us when they do-ed.), it suddenly looks like a true enterprise solution."

So, does any of this mean that the ECM system has to come from a single source? "No," insists Bill Seawick, senior vice president and chief marketing officer at Interwoven. "ECM can mean multiple vendors forming an infrastructure for managing unstructured data. The more interesting facet, then, is not buying the infrastructure, but buying solutions for business problems and then relying on the infrastructure to migrate those solutions from department to department. I challenge anyone to show me a Global 1000 company that uses one vendor and one toolkit. Yes, there's a trend to consolidate down to a smaller number of vendors, but there are very few companies that can do it today."

Richard Holling, director, insurance segment, financial services at HewlettPackard, agrees that the sentiment is accurate...just a little too soon perhaps: "In the past, content management solutions were bought to solve separate problems-one for the portal, one for the claims department, another for records management...IT departments are now recognizing that's crazy, and we ought to be deploying the same software for everybody in the same way we standardize on Oracle or SQL databases, or on UNIX or Windows.

"That said, I'm not sure it's actually happening yet in terms of buying behavior. They want their purchases to fit into an overall content management strategy, but they haven't made the full commitment to go with one of the larger, more comprehensive software vendors and thereby give up some of the functionality they might get from the niche vendors. They want both; to deploy enterprisewide but to also have the best product for the individual task."

John Page, executive vice president, global services for Stellent, has a slightly different take: "The excellent companies think it's silly to have five different vendors, and pay maintenance on five different applications. The same piece of content, perhaps created for a transaction, then has to be shared out to a website, then saved as a record...having three different applications to do all that has created a lot of integration problems and excess cost. The excellent companies want to know which strategy fits their overall needs as an enterprise, and get rid of the other ones."

And so the "monolithic vs. interoperable" debate is on: Siderean's Stevens says, "Our customers want to use the silos of information they have in an interoperable way. They don't want to replace systems; some of them can't be replaced, and some of them are outside the organization. Remember: the definition of enterprise content includes the spanning of enterprises. In government, agencies need to share information, but they have no control over what systems are used in the other agencies."

Oracle's Buchheim acknowledges the realities of the siloed organization, but points out a crucial missing element: "In an insurance company, the claims processing application is managed by a single group, so it is departmental. But the fact is, those claims need to be made available to virtually everyone in the organization at some point or another," he says. "Even though there's not always a lot of crossover, being able to manage your HR records in the same environment as your financial content turns out to be a way to leverage technology. It touches everyone who needs it...they are able to leverage the technology that might not be specifically related to the business application. The cost advantage, and the compliance and risk management advantages, of using the same technology are great."

Xerox DocuShare's Murphy has yet another perspective: "The drivers for ECM are the legal issues that arise from having a heterogeneous mix of applications that have been implemented at the department level in the past. The demand for enterprise content management has escalated up to the CIO/CEO level because of the legacy of multiple applications running that are not connected or under any corporate governance. It has more to do with risk than the need to standardize on any particular tool," Murphy insists.

Why ECM Works...or Does It?

It may, at this point, sound self-serving for these vendor dudes to be talking about the marvels of ECM. But they make the case: "Enterprise data drives enterprise processes," declares Siderean's Anderson. "Navigating internal and external data gives organizations the ability to execute business in a more rapid and efficient manner." He argues that combining internal business data with external information from all over the place, presented in a clear "visualized" format, can discover previously unrecognized relationships.

Here's a lurid example: A bank's customer is a high-net-worth individual who happens to work in the entertainment industry. An RSS feed from the Web reports that she's announcing her divorce. The blogosphere is abuzz with rumors that her low-life golddigger husband has been cheating on her. Armed with that much juicy information, the bank can make better business decisions. Creepy example, but you get the idea.

Yet, as has already been pointed out, what is a good idea may not be the norm in the marketplace...yet. Stellent's Page describes a typical scenario: "Within the top 100 of IT organizations, there is a subset that has figured out that ECM is a strategy they need to have enterprisewide. They base their evaluations NOT on 'who has the best records management?' or 'who has the best Web publishing?' but instead on 'who has the capabilities across many aspects of content management?' But when it comes time to buy a license, they only sign up for the initial solution that's first off the block," he says. They're looking for proof, I say. "Yes, you need to successfully deploy before they'll expand the license to other departments. Evaluations are based on enterprise requirements; buying trends are still more directly focused," says Page.

HP's Moiling puts it this way: "The 'talk' is to deploy enterprisewide, but I haven't actually seen anyone do it yet. In the excellent companies, there is no IT/business divide anymore. IT departments are no longer in the habit of saying 'no, you can't do that.' Now. they're saying 'You can do whatever you like-just remember there's a cost involved.' The department's decision is balanced against IT's budget, and they are told 'you can't afford that. But if you'll settle for 80% of the functionality, you can have it tomorrow."'

Interwoven's Seawick insists that "CIOs are not buying infrastructure. They definitely want to standardize on platforms that are leveragable across business units, but the trend in ECM is coming from the lines of business who are trying to solve their problems, which in turn migrate from business unit to business unit through the CIO." Seawick continues: "This trend of using unstructured data to solve business problems is a new frontier. Structured applications have hit the wall; automating the structured data from supply chain, HR, financials and CRM has gone as far as it can."

The "who owns it?" question struck a chord with several of the interview subjects, but Richard Moiling from HP might have put it best: "IT is not a standalone operation anymore. If the CIO is not on the corporate board, he or she is very close to the board. There's been considerable work to put in place structures for the IT department to work closely with the business leaders-they either have business liaison managers in the IT department, or IT liaisons in the various business departments. It's their jobs to make sure business requirements are translated to the IT department.

"Of course," he adds, "in smaller organizations, there are still those who see IT as a pain in the neck-something they have to do."

Is ECM something you have to do? Read on and decide for yourself.