With the quick pace of the information age, data management is a continuous and thorny problem. We are all drowning in data. It exists in multiple locations and formats. Some of it overlaps or duplicates other information, some is unique, some is static, and some changes periodically. The organization that can codify, manage, and repurpose information effectively will have a tremendous competitive advantage.

The terms content management and single sourcing are often used interchangeably, but they mean very different things. Content management refers to a system for categorizing information to minimize duplication and maximize reuse. Single sourcing, however, is based on the premise that you can develop discrete sets of information once and use them in many places. For true single sourcing, any time the original content changes, it automatically

updates everywhere.

Intelligent content management usually precedes any attempt at single sourcing and is difficult to achieve because it is similar to the paperless office. I have yet to see a business of any size achieve this goal - categorizing and managing content is an ambitious task in itself.

In a workshop I teach to introduce the concepts of content management, I use a simple contact list to show the complexity of the problem. I ask the group how they would organize and maintain a dynamic list of contact information. The ensuing discussion reveals many of the underlying assumptions. To address the content management, organizations must ask:

  • Who uses the data?
  • Who maintains the data?
  • How frequently does data change?
  • How is data distributed?
  • What software programs currently house the data?
  • Is new software needed to manage the data?

I'm willing to bet that most people have at least two or more sources they use to store contact information, such as a printed or online corporate directory, an email contact list, an electronic or paper-based personal list, a cell phone directory, and a printout of emergency contacts on the refrigerator.

Self starter

The more we try to consolidate information in a single area, the more we run up against fundamental limitations of knowledge, time, and tools. Not surprisingly, organizations often try to address content management with tools. There are a number of software packages - from shareware to multi-million-dollar enterprise resource planning solutions - that claim to offer solutions. However, any software solution will fail without significant investment in analysis and categorization.

Based on experience, the best approach is to start with a small set of data with important but limited reuse within and outside the organization. For example, you might start with the data required for an existing product - the information required for maintenance, marketing, training, and support.

Starting with an existing product removes the pressure of deadlines and helps you focus on structure. The challenge is to understand how to label and store relevant data to best serve the needs of all the people who use, create, and maintain it, and to do this in a way that is scalable for your organization.

You need to identify who touches the data in every place along the chain, and determine how they use it and refer to it. The next step is to define naming conventions and methodologies. How does content arrive? What format is it in? How often does it change? How should you label types of data for consistency?

Exploring these questions in detail will help you find a solution that fits the needs of your organization before you spend money on a vendor. Defining the structure, data flow, and categories of data is essential to achieving effective content management. If you get this process right, you may find that you need defined processes, not an investment in products.

Don't do it alone

A team of individuals who combine content management expertise with in-depth knowledge of your organization and how it works should be responsible for creating a strategy. In-house staff might have the organizational knowledge but not the content management background. So combining consulting and in-house expertise is often the only structure that works for the organization.

Organizations that attempt to handle the initiative on their own may encounter road blocks. For example, a global technology corporation that embarked on a content management initiative and created an in-house team with representatives from each region, was plagued by overlapping definitions and political infighting.

When content management consultants came on board, their first job was to show the team content structures that might work. This diffused tensions and helped the team to regroup and work together. Had consultants with the right expertise been part of the team from the beginning, the company would have saved months of time, a significant investment in the project, and unnecessary infighting.

Look before buying

It's much easier to make a case for buying a product than devoting the time needed to analyze and develop an underlying structure for content management. This problem is exacerbated by the fact that content management vendors downplay the effort required and focus on the features of their particular solution. With a little research, it's possible to make a case for the necessary investment in analysis and structure.

Even if you don't have support and funding for a major content management project, it is still possible to organize content. In the field of technical writing, the integration of FrameMaker, XML, and DITA (Darwin Information Typing Architecture) provides a basic toolset that allows information developers to design information in tagged, reusable chunks for semi-automated assembly into multiple documents.

For example, a writer might create a corporate or product overview that can be used on a website, in marketing material, and in a variety of manuals. Adding training to the mix complicates the problem, however, because these tools are inappropriate for instructional design.

Collaboration

For training, it is especially important to define the sources of data, both in terms of who creates it and the software in which it exists. Working together with information designers, it is possible to leverage content management efforts that may already exist or be in the works. For example, if technical writers create chunks of content about a piece of equipment, instructional designers can use the information to train an audience on how to operate the equipment.

To develop a training outline, you need to understand how writers are labeling overviews, creating organizational procedures, and troubleshooting. Even if you need to convert data, you can leverage existing efforts and minimize the need to redevelop existing content. This is a major step on the road to content management and may be a reasonable first step for your organization.

At a global equipment manufacturer in San Jose, California, technical writers were producing 1,500-page support manuals in FrameMaker that were rigorously edited. The problem was that these guides were obsolete before delivery was made to global service technicians.

The training team worked with the technicians and writers to craft a solution. Using XML, the training team developed a web portal for appropriate documentation to appear in relevant pop-up windows based on graphic overviews of equipment. This meant that the writers could revise specific modules without reissuing the entire manual. The training team could add training procedures, videos, and other job aids through the portal, which technicians could then download to their laptops as needed.

The content management portion consisted of writers and trainers working together to define, organize, and label text segments for conversion to XML. The results were immediate and rewarding; the team continuously adds features to the in-house portal. It helps to think of content management as an ongoing process, not an end result.

If you start with a manageable set of objectives, you will make progress. The trainer's old standby is that a needs analysis is always the first place to start. We all have that tool and can use it to define a reasonable order for the training environment.