Purple pitches put portals in peril

Purple pitches put portals in peril

Portal is a buzzword, but that doesn't mean it should be dismissed out of hand.

By Paul Montgomery

Users of information management technologies could be forgiven for experiencing a bit of buzzword burn-out. Just when you've got your head around the difficult concept of knowledge management, vendors start pushing the "enterprise information portal" concept at you like it's the panacea for all your ills.

No doubt you would have heard all of the latest tricks vendors are using to woo the punters. "This widget is going to be the next big thing to build on the work you've done on ERP suites!" some will tell you. "After you've finished Y2K, everyone's going to get this doohickey!" others will plead. "This thingumajig is the enabler of your knowledge platform!" is the cry from the rest.

Here's the bottom line. A corporate portal is a way for you to get rid of your IT staff, or at least make them less important.

Just like knowledge management, many vendors are claiming to offer products in the loosely-defined portal space, without actually addressing many of the areas which a true portal is supposed to. Potential customers would be right to fold their arms in a presentation and say in a sceptical tone (as apparently has been happening), "Say that again, and this time, no buzzwords!"

As with any new technology, the problem is that it sounds fabulous but to justify expenditure on it there has to be a business case, almost invariably tied to a specific benefit or problem solved. Like the paperless office, the portal could become a nice idea that few ever take seriously.

So what is the business driver? Here's the bottom line. A corporate portal is a way for you to get rid of your IT staff, or at least make them less important to internal processes. How does this wondrous feat come to pass? you ask. By making users less reliant on the closed systems operated by IT staff. The current power balance between IT departments and users within an organisation is coloured by the reliance on IT staff to operate the applications and infrastructure around which normal employees base their working days. The portal shifts this in favour of the users, because (theoretically) the interface to every data source is integrated. The Web becomes the "look and feel" for every application, and every application is able to be delivered to any user on a needs basis, with personalisation features to help the user take control.

That's the theory. In truth, no vendor has a product that can do everything a portal is supposed to do. This is partly because the market is immature, and vendors have only just discovered the field themselves. It is also because a portal is complex enough to need at least three different IT disciplines to be mastered to deliver the promised functionality. Document management, represented by PC DOCS, Insight Technologies, Tower Technology, Open Text and others, is an essential part. Business intelligence, from vendors like Cognos, Information Advantage, Information Builders, and Sqribe, is also necessary for high-level decision making. Search engines, like those of Autonomy, Excalibur, Verity and Yahoo!, have to be included to enable the masses of corporate data to be used efficiently. Access to mainframe applications is another important component, offered by companies such as Esker.

There are other vital areas of expertise, like interface design, personalisation and collaboration, but the lesson from all of this should be clear by now. At least in the short term, there is not a complete off-the-shelf solution for the enterprise information portal, and users who want all of the functionality will have to do it themselves - most likely, building on top of one or more packaged products.

This is not a bad thing, although very few are going to be that adventurous. Those organisations which do stick their necks out for a lengthy development project could grab the opportunity for competitive advantage over less agile rivals, but realistically you've got at least six months before it's safe, post-2000, to implement. Then the true worth of the portal will be found out.