Content management systems: Don't automate the misery

Organizations of every size are attempting to get a handle on their content generation, management, and publishing systems. This trend toward business process re-engineering (BPR) of content management is largely the result of an outsized proliferation of Web pages, intranet sites, and electronic communications strategies adopted by organizations, their partners, and customers.

Sadly, few organizations have seen much good come of content-management BPR initiatives so far. Of the many reasons for these failures, one stands out: these BPR initiatives—and the systems they spawn—are focused on realizing organizational objectives without sufficient regard for the context, habits, and goals of the people who will actually use the system. These new technology solutions are intended to create efficiencies, but they actually prevent people from achieving their objectives, which generally have to do with reducing hassle and ensuring their own personal effectiveness.

A typical content management system will include a meta-tagging normalization function to make sure attributes and information associated with the content remain consistent and thus easy to manage. But these functions are usually so difficult to use and foreign—particularly in collaborative environments—that few people have patience for them. Since people are more concerned with getting content in the hands of their audience than they are with meta-tagging normalization, they often skip this key step.

And so an ugly cycle arises in which organizational objectives thwart human initiative, and human behavior thwarts organizational objectives. Managers are then placed in the unenviable position of threatening people with their jobs if they fail to adopt the outlandish practices that are merely artifacts of a senseless system.

Look before you leap

The ultimate aim of the new generation of content management systems is generally the promotion of customer sales and satisfaction, along with an overall decrease in content management expenses. But what's rotten at the heart of the planning and development of these systems is pretty much what cripples stand-alone software application development: before understanding the problem, organizations select a solution. Rather than assessing the organizational objectives and human goals, it is common for business process re-engineers to go out and buy a system of technologies that sparkles with the promise of precision and control, purporting to solve all problems, so long as they are generic.

Part of the problem with this approach is that there is no such thing as a generic problem. Consequently, people spend a lot of cycles adapting the system to their needs, or (more commonly) they are forced to adapt their behavior to the unyielding system. This is a shame, for a system that disregards people will itself be disregarded by people. If the new system requires as much duct tape and bailing wire as the one that preceded it, the efficiencies hoped for are squandered and everybody loses.

Don't automate the misery!

So how to restore sanity? Business process engineers face a daunting set of challenges as they design, engineer, roll out, and monitor the effectiveness of new content management systems. To succeed, they must create coherent sets of roles, transactions, workflows, business practices, and technologies. They need to make allowances for reasonable differences in business practices throughout the enterprise. They have to garner support from line managers. They need migration plans that guide the organization in its adoption of the new system while minimizing strife. Finally, they must have visibility into and control over the system itself so it can change in accord with changing imperatives.

Everybody knows their existing systems are painful and perversely complex, so a palpable fear hovers among many that any new system will simply render into silicon the poor practices people have been trying to shed all along. In essence, nobody wants to automate the misery. But they can't help themselves. In fact, they pile it on. That is why it is particularly important for new content-management systems to understand and appreciate the often bewilderingly heterogeneous practices people use to manage content. Without an in-depth investigation and analysis, it is difficult to say whether any single practice was formulated by legitimate business contexts and goals, inadequate organizational or technical resources, or the eccentricities of highly-caffeinated individuals with too much work and too little time. Consequently, it is difficult to say whether it is worthwhile or counter-productive to incorporate any given process into the new system.

To make that distinction, you need to leverage the experience, judgment, and initiative of the people the system is ultimately intended to serve. Before choosing a solution, it is critical to first ascertain the habits, expectations, and goals of people—as well as the objectives of the organization—so that, ultimately, the system designed reconciles these when they come into conflict. One method is to develop individual and organizational personas—fictional representative users that are essentially inexpensive stand-ins for the real thing—and put them through scenarios emblematic of how they would use the system for, let's say, assigning meta-tags to content. These scenarios can then generate well-defined catalogs of needs that, in turn, can serve as the foundation for the new content-management system's most-favored practices and functional elements. The key at this point is to envision an efficient system of procedures and technologies that gives humans the guidance they need to serve the organization, while providing them the room they need to exercise their best judgment.

When building a content-management system, it's essential that it not only look good on paper, but that it avoids institutionalizing any dysfunctional dynamics in the organization as well. The best way to do that is to model the system ahead of time with organizational and individual personas, whose goals will determine the shape of the system. Only in this way can we stop thwarting one another.