I3 UPDATE / Entovation International News

a free monthly briefing on the knowledge agenda

No. 58 February 2002





David J. Skyrme


David Skyrme Associates


Contents - Readers Reply - Knowledge Digest


What Price Enterprise Systems?

David J. Skyrme

Something that never ceases to amaze me is the apparently high price that is charged for so-called 'enterprise' systems. Over the last few years there has been a convergence of various kinds of software into suites that are variously called portal systems, content management systems and knowledge management suites. Coming from different heritages (Search, web development, document management etc.) these suites now boast an impressive array of components:

  • intranet / portal: browser access into wide range of information and applications
  • document management: multiple input formats, view original
  • content management: creation of blocks for multiple locations; version control
  • communities: see who is accessing particular documents; discussion space
  • (expertise) directory: who knows what; expertise; profiles
  • classification: automatic classification based on AI, but open to users to alter i.e. semi-automatic
  • search and retrieval: ability to index multiple formats; multiple methods: natural language, Boolean, automatic root, proximity, numeric, term weighted, thesaurus integration, search by (metadata) field and type, concept searching ("more like this")
  • agents: automatic searching for similar information
  • workflow: build rules, visual aids, administrator controls
  • alerts: alert user when new items for users profile (or changes, user ability to set parameters)
  • help-desk logging: keep track of requests

To make these components all work together, vendors build them with standard interfaces e.g. to ODBC compliant databases, and usually provide a wide range of import/export facilities. Behind the scenes they convert various data formats, generate indexes and hold metadata repositories. They are enterprise-wide when connected though a corporate network and users from any location can access the functions and information for which they have access privileges.

However, even a basic implementation of such an enterprise suite is likely to cost $250,000, and the typical bill where several thousand employees are users is likely to be several millions. In the context of a KM programme which might cost several times that, and where potential savings can exceed $10 million a year, it might seem small beer. But are these enterprise systems really worth the price premiums they command?

You Get What You Pay For

One argument is that you get what you pay for:

  • Industrial strength software - less errors and ease of recovery when they occur
  • Ready-to-run 'out of the box' - at least for standard defaults
  • Wide range of access controls - providing high levels of security and authentication
  • Built-in management features - user and usage reports, audit trails etc.
  • Coexists and works with standard office suites
  • Extensive support and back-up

What is often not said, and I suspect accounts for a significant portion of the price, is that you also pay for:

  • Expensive corporate account sales people who sell these systems in face-to-face mode
  • A lot of hand-holding; after all the wider a system is dispersed around an enterprise, the more likely there are novice users
  • High cost of development spread over a relatively low volume of sales
  • Interfacing costs - being an enterprise system, users expect it to work with existing enterprise systems, and a wide variety of software that is found in an enterprise
  • The 'brand' - if you are market leader, then you can command a premium, because you can tell your boss that other users - well-known large corporations would not make mistakes, would they? (the modern version of "you never get blamed for buying IBM").

However, does everyone need all these extras, and what are they really worth to you?

The Alternative View

My own experience, especially in the field of CM/KM software which is evolving fast, is that many so called 'industrial strength' systems are as prone to problems as any other type of software once you stretch them beyond very basic functions. Problems with interfaces, upgrades to certain modules that affect older modules you have installed, idiosyncratic data entry inconsistent with commonly used methods, and many other things that can go wrong do. The problems are compounded by the fact that you may have used a systems integrator, thereby opening up the opportunity for two vendors to blame each other for problems. Also, your route to the developers is often far from direct - going through layers of software support specialists who probably do not use the system as intensively as you do.

Here are some other things to think about:

  • The enterprise system may not be the best for the functions you require - ideally you might want best in breed for your critical functions (e.g. search, communities); how easy is it to plug in alternative functional modules from other vendors?
  • Many enterprise systems are inflexible about what you can change. It may be difficult to customise it to your needs (and if you do, you may then have to recustomise it when the base product is upgraded)
  • The price of proposed solutions to the same functional specification can vary by a factor of three or more
  • Licence prices may be per seat or per server. Often you pay for more licences than you can really take advantage of, yet if you under-estimate you can rarely pay for the incremental access you need on a pay-as-you go basis
  • There is an impressive amount of Open Source software that can with the right functions provide much of the functionality of a basic CM system (e.g. MySQL and PHP scripts).

If you shop around, and are prepared to take a risk with a less known supplier, or one that specializes in a given functionality niche, you really can find solutions (albeit not as comprehensive) that are a fraction of the typical enterprise solution. To give you two examples:

Ektron (http://www.ektron.com) - provides a fairly robust content management system with workflow, database etc. that works with Microsoft ASP or Cold Fusion server for only $495. Even their enterprise version costs only some $25,000 compared to enterprise CM systems that perhaps have only 10 per cent more functionality and cost $150,000 plus.

Companion (http://www.companion.nl) - if you simply want to build web pages quicker and smarter, then this relatively unknown Dutch product allows you to build pages from components, either static pages in your own PC or dynamic ones on a server from an Access database. Price: $250.

So Is The Difference Worth It?

Although many corporate buyers may have ambitions to have enterprise-wide software, the reality is that many enterprises are pockets of smaller (team and division) initiatives. Often, it might just make sense to experiment with low cost interoperable components and grow your enterprise systems from the bottom up. Also a low cost solution for a particular function, such as Sitescape for communities, may be a better investment for the enterprise than the suite solution that does everything averagely well but is best at nothing.

Don't get me wrong. I'm not saying do not invest in an enterprise solution. If it does what you want with minimal tailoring and the cost per person is competitive, it may well be the best for you. However, do think beyond the vendor hype. I've seen some very expensive 'enterprise solutions' that are difficult to use and adapt (considerably adding to user costs), yet at the same time some very effective KM implementations done with much lower cost software (and lower admistrator/user costs).

Whenever you are looking for KM software, just make sure you are really clear about your requirements and options. Be sure about what you are getting for your money, what the associated development and people costs are, the costs of upgrades and annual licencing. Think of ease-of-use and what the training and support bill for users might be. And for your the sake of your company's bank balance, just make sure you have explored the lower cost options.

Whichever route you choose, the system is only one part of the equation. It is the people, processes and management of the knowledge flows that will determine its ultimate success, and whether that success is local or enterprise-wide.

Email: David J. Skyrme

© Copyright, 2002. David Skyrme Associates Limited and Authors - All rights reserved.

I3 UPDATE / ENTOVATION International News is a joint publication of David Skyrme Associates Limited and ENTOVATION International Limited - providers of trends analysis, strategic advice and workshops on knowledge management and knowledge innovation®

® Knowledge Innovation is a registered trademark of ENTOVATION International.






Customers: a new twist on knowledge management

Dot com winners and losers

Virtual teaming and virtual organizations: 25 principles of proven practice

Measurement myopia; those who measure and those who act

Portal power: gateways or trapdoors?

Creativity is not innovation

Virtual trust

China: accepting the knowledge challenge

Innovation action for Europe