Related Links

Web Seminars

Rich would like to thank Steven Piotrowski, senior consultant, Deloitte Consulting LLP, for contributing to this month's column.

In April 2006, I wrote a column on enterprise content management (ECM) in which I defined the concept of ECM and talked a bit about the ECM lifecycle. The response I've received on that column over the past year has been enormous, and most people I've talked with have asked for more in-depth coverage of the topic. I'm only too happy to accommodate them!

Before delving too deeply into ways to manage and organize the ECM lifecycle, however, we need a working definition of ECM, sort of a primer to establish what ECM is - and isn't. First, ECM is not knowledge management, which is the process by which companies generate value from their intellectual assets through the efficient sharing of knowledge enterprise-wide. It's also not document management, which is the effective storage and retrieval of documents - period. Of course document management is a part of ECM, but it's not the whole.

ECM is the integration of various technologies and processes to manage enterprise content from conception through deployment. Usually, ECM is a corporate initiative to identify, capture and disseminate unstructured information throughout the company. This information can include paper transaction and other records, emails, Web pages, forms, photos and videos, digital certificates, faxes, word processing documents and spreadsheets, and any other paper documents the company might generate.

Typically, ECM includes an effort to organize this content in a meaningful way by creating taxonomies and metadata definitions. It also includes efforts to deliver content through various channels and search mechanisms. It's the effort to organize the content in a "meaningful" way that I'll be discussing in this column.

The ECM Lifecycle

Before discussing organization, however, let's summarize the steps in the ECM lifecycle. They are:

• Creation
• Review
• Storage
• Distribution
• Archival/destruction

 Figure 1: graphically illustrates the ECM lifecycle.

Creation is how content is born. It's created through new activities, revisions of existing content and acquisition from a variety of sources. The result is massively distributed content creation.

The next step is review. The review process is driven by various relevant business areas to validate content contributions by using business rules that help contributors create meaningful content through checks that provide accuracy.

Next up in the lifecycle is storage. Approved content is stored in a central repository where it is organized and codified for broad enterprise use and distribution - which is the next step in the lifecycle.

In the distribution step, content is associated with a "presentation layer" for access and distribution. Formatting makes the content conform to the expectations and demands of the user community or distribution channel.

The final step in the ECM lifecycle is archival or destruction. Content is archived or destroyed based on business rules. Content remains in archives, allowing future workers to benefit from past work. It should be destroyed when it has outlived its usefulness. How well companies organize and manage their ECM lifecycle will play a large part in how useful enterprise content is to the company on an ongoing basis.

Taxonomies - Organizing ECM Enterprise-Wide

Effective organization of enterprise content begins with developing a workable, efficient, effective taxonomy system so that the content can be easily and accurately stored, parsed, retrieved and disseminated. Taxonomy is just a 50-cent word for a classification methodology that's used according to a defined system that facilitates information retrieval and determination of information relevance. Taxonomies are typically organized into mutually exclusive groups and subgroups of information.

There are several inherent characteristics that shape effective taxonomies. The first is a well-defined classification schema. A good schema facilitates straightforward identification, arrangement, and management of enterprise content. An effective classification schema might be one that is function-based with multiple levels, such as function, subfunction and activity. An example of a function-based schema could come from the human resources function. The function is human resources; the sub-function would be recruitment; the activity might be job posting. All content - whether it's online or paper based, would be classified and stored under this schema.

An effective taxonomy also has a well controlled vocabulary and thesaurus. This is simply a list of preferred and acceptable variant terms, with relationships - both hierarchical and associative - defined for the company so that the end user can adequately interpret them. Such definitions might include the terms: account, client, group and customer.

Another characteristic of an effective taxonomy is the presence of robust indices that facilitate content browsing by category and personalized views that are meaningful to users. These indices provide a way for users to view their desired content based on subject, or other categories, in the classification schema I mentioned above. For a multinational company, an example would be a set of indices that facilitated sorting content by country, then division, then department.

The final characteristic of an effective taxonomy is a set of metadata standards. Metadata contains descriptive information about a piece of content - i.e., data about data. You can think of metadata standards as a sort of library card catalog (for those of us old enough to remember them!) that describes the enterprise content in a uniform manner that makes it easy to find. An example of metadata terms that describe a particular piece of content might be: author, document title, creation date, and subject. Each of these terms would be defined uniformly throughout the company.

Taxonomy Types and Their Potential Benefits

Now that we know what makes an effective taxonomy work, let's move to a discussion of the different types of taxonomies. There are myriad types of taxonomies out there, but there are three types that are probably the most commonly used:

• Subject based
• Organization based
• Function based