Keys to Successful Content Management: Avoiding the Traps

Home/Publications/Best Practices Newsletter/2004 – Best Practices Newsletter/Keys to Successful Content Management: Avoiding the Traps


April 2004

Keys to Successful Content Management: Avoiding the Traps

CIDMIconNewsletter Scott Wolff, Manager, Content Management and Integration, Imaging and Printing Systems, Hewlett-Packard Company

With the ever-increasing industry challenges to reduce cost and time to market, many businesses are turning to content management to reap the benefits of single-source content management solutions. Among the promised benefits are reuse, version control, and automation of workflow and desktop publishing. With true single-source content management, content can be written once and reused in many different publications. However, reusing information requires that your plans address the information design, not just the systems used to manage content. In practice, most content management solutions fail before the first application is even purchased. How do you beat the odds and build a successful solution? Perhaps one of the most basic issues is the desire for simple solutions to complex problems. To beat the odds and build a great solution, you need knowledge, expertise, experience, and, most of all, commitment.

Executives are becoming acquainted with XML, they’re hearing the benefits, and they want their businesses to reap the rewards. However, there are no off-the-shelf solutions to content management. Success starts with educating your sponsors and executives, not only on benefits and return on investment but also on setting realistic expectations for cost and timetables. The rewards are achieved only at substantial cost and through directed strategic investment. Executives need to understand the hurdles, and you need to put together a series of stepwise solutions, where each success builds on the previous one.

Because these solutions are complex, there are many traps to fall into along the way. Any one of these traps can result in the complete failure of your program and the loss of substantial investments. In this article, I share with you a few of the most notorious traps to avoid.

Let us start right from the early stages of any content management program. Here are eight potential mistakes you should avoid. By avoiding these mistakes, you will be on the path to success.

  • Trap 1: Believing that you have the right skills and team, when you don’t.
  • Trap 2: Starting your program by locating a content management system (CMS) vendor.
  • Trap 3: Starting your design by defining the system.
  • Trap 4: Using your CMS vendor to manage the solution design, creation, and integration.
  • Trap 5: Believing that one physical storage solution can meet your entire business’s content management requirements.
  • Trap 6: Choosing applications that require proprietary information embedded in the content itself to function properly.
  • Trap 7: Not performance testing the solution before purchasing it.
  • Trap 8: Attempting to build the entire solution from the start in one pass.

Trap 1: Believing that you have the right skills and team, when you don’t

Every great system starts with a great team. You need experienced system planners, business subject matter experts, expert information analysts, XML template authors, system integration experts, IT support functions, trainers, a supportive user base, and solid sponsors. Content management programs tend to grow either from point solutions conceived by end users or from IT organizations, which often have a mandate to deliver a solution on behalf of their business. Either way, business teams and IT teams often lack the experience to make sound decisions. The key is having the right people engaged at each step of the program. You need people who fully understand the technology and the business requirements. Your program simply cannot have enough experience, and it takes a number of different disciplines all working together to build a good solution. Before you take your first step, seek out expertise. Find the people who own the publications and information for which you are creating the solution. These are the people who understand what has to be done and why it gets done each day. If your team is deficient in a discipline, such as information design or template development, look to external expertise: hire a consultant. Build a solid team with the right skill sets and depth of experience to succeed.

Trap 2: Starting your program by locating a content management system vendor

Often businesses start out by identifying or discovering a particular content management application. The first thing you should realize is that the less experience you have with the particular technology, the more likely you are to get attached to a specific vendor too early in your development program. It is an easy mistake to make, since it seems like a logical place to start: finding an application vendor who can help you solve this problem. But every application has a different set of strengths and weaknesses. Before you ever choose a vendor, you need to understand exactly what business problem you are attempting to solve. There are a number of excellent books that can help you ask the right questions to scope your business need. The questions include: What type of reuse are you after? What end deliverable format are you creating? And will external customers dynamically interact with the system? Once your team has defined the business requirements, prioritized their importance, scanned the industry for available solutions, and keyed in on the core set of differentiating application features, you will be able to make a more informed choice.

Trap 3: Starting your design by defining the system

This trap is related to trap 2. If you made the fatal mistake in trap 2, you almost certainly fell into trap 3. Content management is about managing an asset, the content. Once you have a solid business problem and business requirements, the next step is to design the information architecture, not the system. If you started with a system vendor, you started with a system that may or may not meet your business needs. This mistake results in backwards design. Within a short time, you will find yourself designing your business around the system rather than vice versa. This trap is so common it catches almost everyone, even those who sidestepped trap 2. Do the upfront analysis and design the content before you even consider applications and hardware. Understand how you will structure the information before you build a solution to manage it. If you haven’t resolved trap 1, do so now! Trap 3 often results from traps 1 and 2. Information design and analysis expertise is often hard to find or altogether missing.

Trap 4: Using your CMS vendor to manage the solution design, creation, and integration

By now you are probably beginning to see a trend here. Trap 4 is related to traps 2 and 3. Often there is a tendency to have the “information experts” be the same as the application vendor chosen to provide the content management application. You will find that the application vendors often offer information design and integration consulting services. Sometimes your application vendor can do a fairly reasonable job; often, however, they don’t. The reason is simply that there are no off-the-shelf content management solutions. A great content management solution requires the integration of a variety of applications, including authoring, storage, provisioning, formatting, and delivery. Putting the storage application vendor in charge could be a bit like asking the fox to watch the hen house. If you have a well-behaved fox, you might be okay, but the odds don’t favor it. My recommendation is that, unless you have the expertise in house, you choose a separate, independent integration vendor to help you through the application selection and integration process. That way your solution can remain content focused and application independent.

Trap 5: Believing that one physical storage solution can meet your entire business’s content management requirements

I have often heard it said that content is king. Organizations within businesses often struggle over where the content gets stored. Implicit in the where is who owns the content, so it is only natural that medieval organizational thinking often surrounds content management systems. The problem here is that content often goes through elaborate life cycles, wherein a number of different organizations conspire to add value at different stages. The information authorship may be quite different from the downstream publishers and consumers of the content. Effective information design requires a syndication approach to information, in which the asset moves about the organization feeding different needs. A syndication approach usually means that a host of different systems pass the information back and forth, since each system often has a different business purpose and handles information with varying degrees of granularity to support differing publication demands.

Trap 6: Choosing applications that require proprietary information embedded in the content itself to function properly

Proprietary information is a problem that you should avoid as much as possible. The greater the separation of information structure from application and system, the more successful you will be at having kept application proprietary features out of your solution. To obtain the greatest return on your content management assets, you want to make the most of what the open standards technologies such as XML have to offer. Application independence helps protect your investment by allowing you to evolve your solution as technology evolves. Today most applications are obsolete in two to three years. If you allowed vendor-specific features into your design, you trap your content with that particular vendor. My best advice is, don’t do it.

Trap 7: Not performance testing the solution before purchasing it

Every program ever created seems to under-invest in testing. Under-investment can happen because the program schedules are often unrealistic to start with. The rule of thumb here is don’t believe anything until you test it. Believing the application vendor without having at least unit-tested the component to ensure that it works as billed is a mistake. Often design decisions are made regarding technology that is still under development. Most businesses can’t afford the issues associated with new, untried technology. This trap kills programs by taking much longer than expected and then under-delivering on expectations. At this point, sponsorship support can turn to disappointment. Protect your program and your sponsorship. Test, test, test.

Trap 8: Attempting to build the entire solution from the start in one pass

One of the best strategies for building complex systems is to do it stepwise. Your program builds on success, and your architecture is tested at every step of the way. Trap 8 can occur when unrealistic expectations are made for any program. “We have only so many weeks to get this done” can spell doom for any program. Realistic expectations make for successful programs. Transforming your business should be approached thoughtfully. By building in success a step at a time, your sponsors have confidence that their investment dollars are working; and your program will have enough time to develop its content information design, develop its system integration, and manage the organizational changes needed to reap the rewards of a great content management solution. By avoiding these pitfalls and practicing sound, stepwise design methods, you can build a great content management system. The first step is to post a copy of these traps so that your team can beat the odds. Then you can tell me how you beat the odds. CIDMIconNewsletter

About the Author

Scott Wolff B&W cropped7

Scott Wolff
Manager, Content Management and Integration
Imaging and Printing Systems Group
Hewlett-Packard Company

Scott Wolff is manager of a Learning Products Group within HP’s Imaging and Printing Systems Group (IPG). IPG Learning Products teams are responsible for the management and creation of product content for HP LaserJet and Inkjet printers and digital imaging products.

Scott has helped lead the rollout of an XML-based, single-source content management solution to support twelve Learning Products groups across HP. He is currently leading a company project to use standard XSLT style sheets for publishing all HP technical product documentation.

Scott is known as an agent of change and has helped champion efforts to transform HP’s content management methods and systems. He has a bachelor’s degree in Industrial Engineering and master’s degrees in Manufacturing Engineering and Computer Science.