May 2002

This month we provide a summary of the latest trends in content management. Our aim is to help you think not only about individual content management projects, but also about how content management technologies are fundamental to broader IT infrastructure strategies.

Frank Gilbane

Download a complete version of this issue that includes industry news and additional information (PDF)

The Top Ten Trends in Content Management

One of the most popular free downloads on our website is an eighteen month old issue called “What is Content Management?”^1. We think it is still a useful report (naturally!), and recommend it to those new to the subject, but the report lacks the insight of recent developments and is increasingly of historical interest. The content management market has changed dramatically since then, and promises to remain diverse and dynamic. This means there will continue to be animated discussions among analysts and vendors on just what content management is. There are many perfectly reasonable definitions of content management that have been created by very smart people. However, in spite of the title, our earlier report did not propose a crisp definition ready to take on all comers, but described where the term came from and how it was being used at the time.

Our aim is to educate and facilitate rather than promote. So, while we often come up with definitions to help a particular audience understand segments of the product and technology landscape, we have always been more Wittgenstinian in our approach. But don’t worry, you don’t need a refresher course in 20th century philosophy to understand either content management or our point of view – we just mean that it is simply more useful to understand how we (users, vendors, analysts) use the term ‘content management’, than it is to argue over whether definition A is better than definition B. There will always be many definitions and all will be influenced by the definer’s experience and motivations.

So, although we didn’t title this “What is Content Management? – An Update”, we are, in a very real sense laying out what content management today is all about. Each of the trends we discuss below extends our understanding and use of “content management”. In a sense, we are simply reporting how the term has grown in the last eighteen months.

One Megatrend

Because of our view, history is important. In fact, history is essential to understand where content management and content technologies are headed. You can read our earlier report to learn how the term ‘content management’ came into use, but there is a much more fundamental historical trend that explains why content management technologies are becoming the central focus of our use of computing technology.

This “megatrend” is that the primary use of computing technology continues to evolve to be more for communication than for data crunching. We won’t go through the entire history here, but think of the progressive use of computers from numeric calculations, to data processing, to unstructured (text, graphics, etc.) data and document processing, to Web publishing with a steady increase in rich media and metadata, to the incorporation and use of all content types in enterprise applications to better communicate with customers etc.

Content is what we communicate, and effective communication is about as fundamental a business requirement as you can get. The insight that computing technology development is increasingly targeted at communication is neither original, nor difficult to see. What is useful about it is that keeping it in mind will help your predictions about near and long-term technology directions, and therefore your IT strategy, stay on track. It may be completely obvious that richer media will become more prevalent in enterprise applications, but it is all too easy to forget in the heat of, for example, major application integration projects.

As we go through the trends you will see that most of them have to do with the increased emphasis on using computing for communicating.

Ten Trends

Ten is an arbitrary number, and inconveniently outside of the number of items easily kept in short term memory, but it sounds good and each of the trends is important enough to be separately identified. An entire issue could easily be devoted to any one of these trends, but our purpose this month is simply to bring them to your attention and briefly comment on each.

Trend one – emergence of second-generation solutions

Our first trend is the only one not directly related to computing and communicating, but is a typical characteristic of software application market segments.

Many software solutions are first developed as highly customized solutions to meet specific and important business requirements. Development is done either in-house or with the help of a very involved vendor or integrator. Because all development is expensive, the business needs have to be important enough to justify the investment, and are often viewed as “mission critical”. It should be obvious that custom development + mission critical = expensive. If the result is successful, it is worth replicating in some form for other businesses. It takes a little while to hone in on what parts of the overall solution are common across companies, so early software markets are by nature characterized by expensive custom solutions.

As the solution becomes more widely implemented and understood it is natural for a new generation of providers to emerge that focus on the most common and generally important parts of the problem, and provide an “80%” solution that is less customized and therefore less expensive. This second-generation broadens the market, and forces prices to come down. There are many other repercussions and further stages in market evolution such as increased vertical specialization.

What is important to remember about this trend is that the content management market has a healthy number of first and second-generation vendors (and arguably a third, but we’ll ignore that for now), and pricing has come down quite a bit.

Trend two – consolidation

There are two types of consolidation relevant to us. The first is vendor consolidation, and the second is solution consolidation.

Vendor consolidation is a standard part of software market segment life-cycles. Most analysts would say that there is only room for 3-5 major players in a market segment. In the content management market, analysts would include Documentum, Interwoven, Stellent, and Vignette on their short lists, although there are dozens of other vendors. There certainly is a lot of traditional consolidation going on in the content management market, with the four vendors just mentioned, as well as many others including Microsoft, divine, and FileNet, buying companies.

Because of the size and scope of content management, consolidation will take longer, and not be as complete or clean, as usual.

Solution consolidation is a bit more interesting. One example of solution consolidation is what some vendors are referring to as “enterprise content management”, by which they mean management of all types of content (Web content, documents, rich media assets). Another example is the incorporation of search and categorization tools into content management systems. A third, and more complex, example involves the inclusion of some content management capabilities in application servers, e-commerce platforms, e-catalogs, and other enterprise applications (ERP, CRM, etc.). This consolidation will keep all of us guessing for a little while. Solution consolidation is directly related to our next trend…

Trend three – content convergence

In the “What is Content Management?” article mentioned earlier we argued that “content” was actually becoming (in fact, not by prescription) a term used almost synonymously with “information”, and that this was a good thing. We have long needed a term that incorporated both structured data and unstructured data^2.“. Information” could have worked but didn’t for some reason. In any case, while we still mostly use separate systems for managing structured and unstructured data because that is what we have in place, nobody (we hope!) would build an IT infrastructure strategy that didn’t include or facilitate the management of both types of data. The distinction between the two is becoming less and less meaningful. How many business applications do you have that don’t require integrating database data with unstructured content?

One obvious, and very critical, sub-trend to pay attention to, is how the database platform vendors (IBM, Oracle, Microsoft) are rapidly changing their platforms to support structured and unstructured data, as well as XML data, which can encode both. (The database platform vendors call out XML data specifically to make the point they support it and because it is only in some newer applications that XML makes up the majority of the content.) Content management and other vendors (especially of the enterprise kind) that sit on top of these database platforms already have to build applications with integrated data types, will have an easier job as the database platforms role out this capability. The solution landscape will certainly be changing as a result.

Trend four – XML

The reason for XML’s success is that it does not discriminate between structured and unstructured content and is therefore a powerful tool for application and information integration. Content management solutions benefit from this at least as much as other solutions do. Whether XML is used for content storage, or metadata sharing or in more imaginative ways (e.g., RDF), you can count on its continued entrenchment.

Trend five – content enrichment or enhancement

As you will see at the end of the article, we think the addition of tools for the enhancement/enrichment (sorry, we can’t decide which term we prefer) of content will define the next stage of content management. There are multiple aspects to content enrichment:

Increased use of rich media
More extensive use of metadata
More integrated multi-lingual capability
Increased use of linguistic analysis for searching, categorization, etc.

As our report last month pointed out, there has been an enormous increase in development of linguistic analysis tools. Note how important these types of enrichment are for communication.

Trend six – content control

What we mean by this could be considered part of the content enrichment trend. But it is important enough to call out on its own. There are two aspects to content control:

Access & authentication. This is not always a job for content management systems, but content management systems will have to be more involved as we use them to manage more granular content components.
Rules & rights. This is not just about digital rights management (DRM). DRM is only a small part of the more general problem of managing what people can do with content once they have been granted access to it. In business environments, there are many rules and processes that have nothing to do with copyright issues. Our ability to associate rules and workflows with content components based on user or user type (employee, customer, investor, supplier), significantly enhances the utility and therefore value of the content.

Trend seven – multi-channel delivery

The most important driver behind the document management market in the late 80s and early 90s was the need to publish information electronically as well as on paper. This dual need exposed the fact that there were few, if any, management processes in place, just as additional management and technical requirements were becoming necessary. While single-source repositories feeding multiple output channels remained out of reach for most applications, at least people started thinking about more effective ways to manage content and avoid completely redundant and wasteful systems. And of course it is not simply a “dual” need; there are many types of electronic channels (PDAs, kiosks, point-of-sale devices, billboards, etc.)

While the Web was responsible for convincing even the most diehard Luddites that electronic information would become the primary, not the only, focus of IT, it is sometimes shocking to see how far we still have to go. The problem of managing content efficiently for delivery to any of the channels appropriate to the business requirement will continue to drive a lot of activity – this is a pretty basic need, and one the big content management vendors are targeting.

Trend eight – portal interface model

This is not a very exciting trend in spite of the hype surrounding invented portal markets. What is important is that a browser-based interface with a single point of access to a variety of repositories and feeds is simply the current model being shared by all enterprise applications. It might make sense for you to build a portal using your content management system, or your ERP system, or your application server, or with a dedicated portal product. The difficult work is what needs to go on under the covers and relates to…

Trend nine – new information integration architectures

This is a very interesting area. We have complained for years about integration solutions myopically focusing on APIs and application integration and ignoring the arguably more important half of the problem, i.e., integrating the information the applications were developed to work on. In addition to the content integration we discussed earlier, and the work the database platform vendors are doing to facilitate it, there is a lot of other activity by EAI vendors and others aimed at the information integration problem. Among them are two categories of solutions we have run across in the past year.

“Enterprise content integration” is a term used in an article published in the Seybold Report^3. to describe vendors, such as Context Media and Agari, that provide solutions for aggregating and managing content from multiple distributed repositories.

“Enterprise information integration” is being used by a number of other vendors, such as Metatomix, Metamatrix, and Nimble, to describe the same capability. The difference between the two categories is simply that the first is focused more on different media types (e.g., streaming media, photos, HTML) and the second on different data types (e.g., ERP and CRM data and transactions).

In both cases these approaches are mostly managing metadata and links to content stored elsewhere. This is a hard problem as there can be severe performance and schema and data synchronization challenges. Note that although most of these solutions are being deployed for specific applications, this is really an issue that should be considered as part of an IT infrastructure strategy. These types of solutions will increasingly look like…

Trend ten – content services

There is a lot of hype surrounding “Web services”. However, when you peel away all the propaganda, you find new ways of talking about incremental progress on a very good idea (“distributed object computing”, “object-oriented programming”, etc.) that has been around for many years. The incremental progress is a result of a consensus on how to approach the problem and its manifestation in standards like SOAP. Web services will happen slowly and in many guises, but to quote ourselves, “Web services will eventually be revolutionary because they will do for communication between computing applications what the Web has done for communication between applications and humans.”^4.. Note the use of “eventually”.

In terms of content management, we think this trend is important because supporting distributed content should be easier than supporting distributed content plus code. In addition to being easier, much, perhaps most, of the business value may be achievable without distributed code. It is true that the distinction between code and content continues to grow fuzzier, but we still expect to see content services role out sooner than full-fledged Web services. A services approach is the best way to address the information integration problem described in trend 9.

Three Stages of Content Management

To conclude our review of content management trends, it may be helpful to describe the evolution of content management systems to provide some context. “Content management” was originally used by almost everyone as a synonym for “Web content management”, and in fact that was an accurate use of the term. By now businesses have realized solving the Web content management problem in isolation often adds cost without solving business problems or generating enough new revenue to cover the costs. “Enterprise content management” is most usefully seen as a term used to portray solutions more integrated into other enterprise applications. Although most businesses have little real integration in place, they, and their software suppliers, are already looking beyond storage, workflow, and integration to how to enrich and enhance their information to increase its communication value.

Figure 1. Three stages of content management

Frank Gilbane
frank@gilbane.com

Notes:

Gilbane Report, Volume 8, Number 8. Available as a free report.
Structured data refers to data that fits neatly in relational tables. Unstructured data typically refers to everything else. There are differences of course between an email message and XML content in a content management system repository.
The Seybold Report, “Enterprise Content Integration: Next Step Beyond DAM?”, January 21, 2002.
“Understanding Web Services“, Gilbane Report, Volume 9, Number 8.

Gilbane Report Vol 10, Num 4: The Top Ten Trends in Content Management