Better Content Management through Information Architecture

Posted by
“To implement a successful content management system, we have to go beyond business process and technology and understand how the organization, as an organism, interacts with and uses its content.”

Everyone understands the business case for Content Management: Organizations drowning in information can’t learn from, act on, or leverage knowledge and resources trapped in assets that already exist. You lose the content’s value if you can’t find it to use it.

To solve problems like these, business often purchases a technology, assuming the former is a feature of the latter. In the content management world, we hear the same kinds of promises from IT stakeholders, again and again:

  • Our developers will provide forms that authors will use to update content online (and every one will live happily ever after).
  • We will use XML (and every one will live happily ever after).
  • We will buy This software from That vendor—off the shelf—and authors will use this to update content. We will customize These options to match our requirements. (And every one will live happily ever after.)

Unfortunately, business often confuses technology for the solution. Forms, XML, and software won’t manage your content. Neither will they help authors create content, nor do they help you leverage content for later use. When Content Management Systems go wrong – and they frequently do – you can end up with terrible nightmare scenarios:

  • Authors write content, and try unsuccessfully to use theCMS.
  • IT updates content in relevant format and uploads for authors to review. The authors make power point presentations with changes and mail them to IT.
  • IT reviews the PowerPoints and uploads the changes.
  • Authors approve the changes, and IT uploads the final version to the site.

To prevent these kinds of scenarios, we try to customize off-the-shelf systems or develop our own, but… IT and business, focused on issues with technology and business process neglect the system-wide ecology that governs how those technologies and processes will be used.

Four crucial factors govern the success of your content management system

Four crucial factors that govern successful CMSs
To implement a successful content management system, we have to go beyond business process and technology and understand how the organization, as an organism, interacts with and uses its content. Four factors are crucial to ensuring an organization can successfully manage its content:

  • Who will interact with the system? Who will create and manage the content? Also, who will need to find and use the content later?
  • What are we managing? What is mission-critical? What kinds of data do we need to manage?
  • How is the system managed? How is the content authored, approved, and managed? How does theCMS enable your business processes?
  • How is the content used? Who will use it, when, and why? How does this integrate with your Information Architecture?

Framing content management in this way helps move the discussion away from business processes and technology. Several familiar methods and tools can help organizations answer these questions and understand how content is created and managed into the system, as well as found and used out of the system.

Who uses your CMS? Authors, approvers, and users (oh my!)

One of the first steps in implementing a CMS should be to identify theCMS users: who will author and publish content? Gather data through user research and modeling.

Scenarios can reveal how the system will really be used, and personas can help business and IT better understand the goals, skill-sets, and motivations that directly impact how successfully the content management system will be used.

The following questions can be useful during user research:

  • What tools are used to author content? At one large research University, instead of composing content using the CMS’s forms, instructors copied and pasted existing content from old syllabi written in Word.
  • How is content authored? At a financial services consultancy, all content was painstakingly crafted by copywriters. In one city council campaign, content appeared as bullet points in emails (or on a whiteboard) that were translated and uploaded by (seemingly) random volunteers.
  • Do the authors and managers understand and use the information architecture? At a large development bank, authors rarely categorized content because they didn’t understand the interface (or its importance). Figure 1 shows a typical content administration screen. Do content authors and managers understand and use the interface?

Figure 1: A typical content administration interface.

Understanding how people use the content is equally important.

  • Who uses the content? In that city council campaign, the website was written for potential voters, but it was mostly used by reporters and precinct captains.
  • How do they use the content? At a large government agency, the system stored versions to help in production of final documents, but writers used stored versions to find examples to copy and paste from.
  • When do they use the content? One music publication published reviews of new releases as a stream of new content, but access was most common for users researching albums long-after they had come out.

    In addition to helping you better understand your users, personas and scenarios can help you test and validate how the effectiveness of theCMS

What are we managing? Handling mission critical content.

What data really needs to be handled by a CMS? In a utopian situation, the database can manage any type of data or content. In reality, it does not work that way. Content management systems usually do not handle all types of data effectively, so you must prioritize critical data and workflows.

  • Where does the content come from? Is it created in-house? Does it arrive in feeds from outside providers? At a large entertainment website, most content came from licensed feeds from outside providers, but in important cases, content was created by onsite editors.
  • What formats do you need to manage? Large enterprises typically manage office documents created, like Word or WordPerfect files. Do you need to manage multi-media files? PDFs? Flash? Implementing search or a decent taxonomy for these formats is a different ball game altogether, and many content management systems simply do not have the ability to manage them efficiently.
  • How often does the content need to be managed? It often does not make financial sense to manage content that is updated on an infrequent basis. Content management systems typically do not have the ability to present brand vehicles – like annual reports – in formats that are aesthetically pleasing and reflect the company’s branding. That job is best left to your creative and marketing team.
  • What content is most important? Critical tasks deserve special attention. Regardless of how frequently critical content tasks must be completed, their importance may require the system allow you to manage their content quickly and flawlessly.

Along those same lines, it may be better to handle a few types of content well, than to handle all types badly.

A broad inventory that lists types of content, sources, and formats can help you understand the potential scope of content you need to manage. Rating the importance of each type of content can help you prioritize what the system must handle, as opposed to what would be nice.

In the example above, handling annual reports might be nice, but managing product descriptions may be imperative.

Along the same lines, task analysis of both common and critical tasks at your organization can further illuminate types of content to focus on.


To implement content management system that really works, business and IT must think beyond internal processes and technology. If your organization keeps in mind the users who will interact with the content as well as the range of types of content, then your CMS is much more likely to succeed.

However, users and content are only the inputs that go into your CMS. Understanding the content lifecycle is the final missing piece. In part two of this series, I’ll examine additional design methods that can help illuminate the early content lifecycle—how the content is managed—as well as the content’s end lifecycle when it finally lands in the hands of end-users.


  1. Well said Masood, this is a key stage in most aspects of business analysis – lifting the Businesses eyes from looking at technology as the solution, to communicating the issues. Your take on this from a content publishing perspective is nicely presented.

    Another aspect which should part of the strategic process is the definition of a communications plan. This should also help the information architecture by defining the nature of the communications, i.e. the dialogues between Author and Reader and some the properties that define what is considered global or local news when dealing with large enterprise projects.

    I look forward to your follow up article, in addition to the lifecycle, will you also be mentioning meta data considerations or internationalisation / localisation considerations?

    Thanks for a good article.

    Richard Marsh

  2. Hi Masood,

    Excellent article! As an industry, I don’t think we can have too many articles that remind organisations of the fundamentals, encouraging them not to get caught up in the technology.

    As I tend to say in workshops: “the CMS won’t solve any of your business problems, it’s just an enabler”. It’s the work around the CMS (publishing processes, content, IA, etc) that really adds the value…


  3. In my experience the biggest issues with content management and information architecture is not that the organization does not understand the concepts you’re talking about. Its more that the development and management of these systems tend to go across the various functional groups and thus leading to fragementation of work and responsibilties. Its like a broken process flow, each looking at its end but never at the entire picture.

    The culture, the organizational structure, the budget etc etc..all these cause the end results of a badly developed and managed content management system and information architecture implementation.

    The vision is always clear and the end result never is.

  4. Adding to the problems of CMS systems is the fact that current popular Web coding languages are embedded in page HTML, like Java or PHP. Sometimes content is interlaced with developer code and how do you protect naive users from accidentally changing or deleting important code, or to even know how to insert content in/around code? This problem has not been solved at all. Even the fact that HTML is potentially insertable into the content is problematic. You might have users who do not know what references the CSS file uses and you may have content that looks inconsistent with the rest of the site.

    Yes you can dumb down CMSes so that code is never shown, but it may so dumbed down that the content can’t change based on other parameters, ie. a widget that scrolls through images related to an article that is embedded only in this current story.

    The interaction between developers, content producers, and product managers through CMS needs further research and thought…

  5. Nice article Masood. I think the problem lies in educating the organizations about the value of User Centered Design. Organizations must consider design as an intrinsic and integral part of all the processes rather than an extrinsic or secondary option. Many organizations still look down upon their design units as a burden rather than a revenue generating center.
    Organizations are full of left brained people who find it easier to embrace technology than design. Vendors convince organizations for buying technology by showing figures; on the other hand design is about value and user delight rather than figures. If we could quantify the value of design successfully for CIOs and CEOs, they would find it easier to embrace design rather than technology and the world would be a much better place to live in.

  6. David Shen, it is possible to separate totally content and code. Authors/editors should only have access to content (create a page, a news line or other information block). Developers may change templates, appearance and logic of information blocks.
    According to inconsistent look of pages. I guess you can filter out most of irrelevant tags on the server side, and do not forget to write a styleguide like this one:

  7. Richard
    Thanks. And your suggestion of defining the communications plan which is anyways a PMI mandated process would be great; especially in this context. In my second part, I talk about the approval and publishing process. the global and the local news is a good example of content categorization that I feel the CMS should provide.

  8. James,
    Thanks. I am fed up of discussions that revolve around technology. What is the RDMS? What is the middle-ware?
    What are the file formats. Is the XML valid? While there is a place for these questions, it should not be at the expense of the crucial factors that are mentioned.

  9. Jay, you are right about organization culture and other factors like budget, but even the vision, even what to expect from a CMS not being clear is one of the fundamental reasons why they fail.

    My point here is why do we expect the authors to know HTML or any other coding language? Authors should focus on writing and not coding. And vice versa.
    You are spot on when you say “Authors/editors should only have access to content (create a page, a news line or other information block)”

    Thanks Praveen. Communicating the value of what we do is a very important part of our job. I use a lot of scenarios in what could go wrong. Step into the shoes of your customer’s customer is what I would do. I do extensive field studies and understand usage scenarios/environmental factors to understand context before presenting to CEOS/ CTOs

  10. There seems to be a lot of overlap here with Mr Veen’s article of several years ago on this topic. I’m not sure if this advances the argument that much, but it’s a nice read none the less. In my opinion, UE/X has far too little to say about content and the management of it.

  11. Nice article. Solid premise, well stated. I also agree with many of the comments regarding organizational challenges to good information architecture. In my experience, not all stakeholders truly understand holistic information architecture, or worse they confuse it with technical design, layout, and simple taxonomy. Information objects in CMS solutions tend to move, grow, change, and evolve. If you take a traditional IT structural approach to setting up taxonomy in a CMS you are bound to fail. Managing information architecture simply has to be user-centric, and is more about understanding user experiences and interactions than it is about portals, channels, pages, and features. Ideally all stakeholders should have a basic understanding of IA, but whoever is responsible for your CMS must be committed to it.

  12. Thanks Eric. The dynamism and evolving nature of information objects is a reality that should be understood, but is often neglected, while planning a content strategy, which I believe is an intrinsic part of information architecture. IA’s should anticipate and plan for the future scalability of content. Only then will your Portals deliver business value. While it would be ideal if all stakeholders understand IA, you are absolutely right that people responsible for CMS should be committed to IA. In fact I say that IA’s should have more stake and be willing to take additional responsibility in content management than what we are currently doing.

  13. Struggling through questions and suggested answers seems to be a growing part of life on the web. I found your article thought provoking. I came to it through a link here:, which is an excellent paper on content management.

    My criticism also comes from reading the namahn article. Mr Namahn points out that IT tends to speak IT and that content management is an IT term, so when one is writing about the subject of content management, one should not use the term. the addage reads that one should never use the term you are defining in the definition.

  14. Thanks mike. You have given me an excellent lesson in CMS. Now am provoked to think and try to redefine the term.. and use the grandma test on it.


Comments are closed.