If you were to survey corporate America on how it defines an enterprise content management (ECM) system, the majority response would be a software tool primarily used by large organizations to create and publish Web content. This article challenges that definition and proposes that ECM is a process that may or may not be supplemented with software tools. In addition, the scope of ECM extends to organizations of all sizes and beyond public Web content to any leverageable informational asset that an enterprise creates, publishes, consumes, and archives.
Before delving into ECM, we must answer two fundamental questions: what is enterprise content and why do we need to manage it? In the broadest sense, enterprise content is an informational asset owned by an enterprise. An article on the corporate Web page is but one drop in the enterprise’s content portfolio bucket, which includes content that is not traditionally managed by ECM software such as strategic plans, proposals, lessons learned, marketing slicks, images, whitepapers, and case studies. If you have ever heard (or asked) the question, “Where is the ______ we did for _____,” you are already down the path to understanding why we need to manage enterprise content.
There are many practical reasons for managing content such as controlling what is published and who can see it, but the fundamental reason is to enable an enterprise to leverage what it “knows,” and more importantly to be able to “know what it knows.” Although the topic of Knowledge Management (KM) is outside the scope of this article, the transfer of knowledge to content and content to knowledge is the overarching concept behind why ECM should be a key component of all enterprises.
The term, “ECM,” is associated with software tools and is typically reserved for medium to large companies due to the significant investment required. But if you expand the context of ECM to be a process that is supplemented by tools, it becomes apparent that enterprises of all sizes already have a system for managing content, formalized or not. Running the gamut from the “wild, wild west” of ECM to highly structured processes backed by automation, the information produced and consumed by these systems is the lifeblood of any organization.
Although there is no right or wrong ECM methodology, there are best practices for content management that can be built upon. “Right” is only measured by how well the system meets the needs of its organization. These needs should be understood before investing in a software tool. It is also critical to have a commitment from key decision makers to improve how information is managed and leveraged continuously. Although it may prove challenging to obtain this commitment, there is a strong case for ECM. By leveraging what it knows, an enterprise can use its experience as a foundation, rather than continually paying others to re-do work from scratch. Not leveraging what you know is like sending monthly payments to the bank when the loan is paid in full—that defies common sense.
The first step toward improving the process of managing information assets in your organization is being able to measure the health of your ECM portfolio against a set of goals. No single set of requirements, or one software tool will meet the needs of all enterprises. Each must establish its own goals, gather requirements, analyze alternatives, and choose between buying and building a tool.
An ECM system exists to facilitate the flow of information throughout the organization. Think of the ECM system as plumbing and information as water flowing through it. Do the pipes in your organization have leaks, bottlenecks or valves that make it difficult for authors to publish content and consumers to find it? Do some departments dig their own information wells and isolate themselves from the enterprise information supply? Analogies aside, a healthy ECM system is one that is an integral part of the business and enables the unrestricted flow of information throughout the enterprise.
Bottlenecks typically occur in the Publish phase of the ECM lifecycle. Authors create content and often encounter an information technology (IT) bottleneck when they attempt to share it with the organization, its partners or customers. This bottleneck results in frustration for the author, increased workload and resource requirements on IT, and reduces the motivation to share information. People have a job to do, and if the ECM system hinders that job, they will avoid or go around it. This results in numerous information wells across the enterprise and cripples information sharing and knowledge transfer. Bottlenecks are not just isolated to IT; however, and can be caused by many factors notwithstanding complexity, bureaucracy, and inefficient processes.
We have established that the flow of information is the overarching goal of a healthy ECM system, but we can dissect that goal into key health indicators found in each phase of the ECM lifecycle. Enterprise content moves through five phases during its lifetime: Create, Describe, Publish, Consume and Archive. The following indicators are not meant as requirements for procuring an automated system, but rather are observable points of a healthy ECM system, automated or not. Using these indicators you can establish measurable goals and embark on an ECM improvement process.
Everyone in your organization has information to share, and the ECM system should facilitate, and encourage this. Treating everyone as a contributor enables people to “give back” and results in communities of interest and a continuous information feedback loop. If only a few can contribute, the content could become biased or narrowly focused. With the ECM open to a diverse author population, content creation must be simple. Authors should be able to work in familiar tools. If it is too complex to create content for the ECM system, the organization will not achieve a balanced perspective on given areas and many people simply won’t share information at all.
Key indicators found in the Create phase are:
Everyone is a potential contributor and should have the ability to author and submit content
Authors should be able to work in familiar tools. The ECM system shouldn’t preclude the author from using their word processor with features such as spelling and grammar checking or force someone to compose a lengthy whitepaper in a text box on a web page
Content creation must be simple. If the process is complex or requires technical intervention, people will be reluctant to share information
Before publishing and sharing content or perhaps even before composing it, the author is aware of certain attributes such as its subject and intended audience. These attributes are referred to as metadata or facets, and enable information consumers to find what they are looking for by browsing or searching. The ECM system must have a strong content typing mechanism that enables information architects to design content types, their facets and associated validation rules. To provide a concrete example, suppose an organization frequently produces case studies and whitepapers about project management. The ECM system would define a case study and a whitepaper content type. The whitepaper type could have facets such as project management lifecycle and target experience level whereas the case study type might have industry and company name. In addition, ECM systems should enforce validation rules to maintain the integrity and consistency of the metadata.
Humans have an innate need to bring order out of chaos. In nature, we categorize by genus and species, and in a brick and mortar library, books are shelved by topic. This organization by category and subcategory is referred to as taxonomy. In traditional library classification schemes such as the Dewey Decimal System, each piece of content has a "correct" place in a hierarchically organized classification system, but there needn’t be a “correct” single-faceted taxonomy. Without the limitations imposed by physical storage, automated ECM systems can dynamically generate multi-faceted taxonomies enabling information to be classified by a series of facets. This enables information consumers to categorize content in a manner that is meaningful to them.
In addition to classifying content by facets, ECM systems should enable authors and content managers to relate content with other content. By creating these relationships, information consumers can be led to other relevant information and take them down paths they may not have anticipated. Automated systems can discover relationships between by analyzing usage, identifying patterns and making recommendations. Most of us are familiar with cross-selling techniques such as “you might also be interested in this,” or “people who purchased this also purchased this.” ECM systems should do with content what retailers do with products. The payoff is in worker productivity and intellectual capital.
Key indicators found in the Describe phase are:
The ability to define content types and associated facets and validation rules
Support for both static and dynamic information taxonomies
Association of relevant content
Once content has been created and described, the next phase involves sharing it with all or part of the organization. The Publish phase is a checkpoint that controls what is published and who can see it. With everyone in your organization as a potential contributor and sensitive information to protect, the ECM system must enforce the publication process. Without process, the ECM system could become a dumping ground rather than a knowledge repository. We use the term workflow to describe the steps that must be taken to publish content. A basic workflow could be:
Author submits content for approval
Content manager reviews content and approves or rejects it
Content is published
The workflow will be different at each organization, and the ECM system should include a flexible mechanism that supports custom workflows.
In addition to controlling the publication process, determining and specifying security requirements is another major step in the Publish phase. This can be accomplished a number of ways, but automated ECM systems typically use role-based security. Each role has a specific set of rights which could include such permissions as “View,” “Edit,” “Delete,” or “Approve.” The ideal ECM system enables roles to be set at the facet or content section level. With this type of granular security, a team could collaborate on a proposal and each person could only view and modify the sections and metadata they are responsible for. The security mechanism should also be flexible to include employees, customers, partners and even anonymous users.
Versioning is an important aspect of the Publication phase. The ECM system should enable the option to store multiple versions of a given content item, but only the approved published version should be available for consumption. There are two types of content versioning:
Publication versioning is the ability to have a single content item with two versions, the version currently being edited and the published version. This type of versioning is an essential feature of an ECM system and prevents the modification of published content while still enabling it to be developed. Modified content should go through a workflow and upon approval, become the published version. Historical versioning is the ability to keep a historical record of changes to content over time. When enabled, this type of versioning should allow content managers to view or rollback to a previous version.
Key indicators found in the Publish phase are:
Customizable workflow to control what is published
Security mechanism that controls permissions to information
Support for publication versioning and the ability to enable historical versioning
The Consume phase has both the longest duration and the most importance to the organization. Through this phase, personal knowledge becomes shared knowledge, and the single most important function of the ECM system is to support this. What would be the sense of gathering, describing and publishing content it if information consumers are unable to find it? To understand the role of the ECM system during this phase, let’s examine how consumers go about locating information.
The methods by which consumers locate information can be divided into the following categories:
Intentional : ask an expert, searching, browsing
Incidental : during an intentional effort, another path to the information presents itself
Accidental : the information is found purely by chance in a way that could not have been predicted
Arguably the top way in which people locate information is to find someone who is regarded to them or by the organization as a subject matter expert. This approach works well in small organizations, but in larger or disparate organizations it is simply not effective. The ECM system does not have the answers to all questions and should enable consumers to locate and communicate with experts. Experts can be self-identified or nominated based off some objective criteria such as participation in working groups, prior experience or current position. If someone executes an intentional search for “Project Management,” the ECM system should not only return whitepapers and case studies, but also people with expert knowledge.
Searching is a pervasive method of locating information, and should be one of the largest investment areas in the ECM system. Information consumers should be able to execute a basic keyword search and be presented with results. In addition to keyword searches, the ECM system should have an advanced search mechanism based around the content types and facets described in the Describe phase. In other words, consumers should be able to make a request such as “show me all case studies where the sector is Public and the customer is Department of Homeland Security. ” Finally, the search should include a learning mechanism that analyzes search patterns and improves future search results.
In addition to asking an expert and searching, consumers also browse to locate information. The ECM system should provide features to support this, and one mechanism is a Library View. A library view can be thought of as a virtual bookshelf and presents a single information taxonomy. Examples of a library view can be seen at http://msdn.microsoft.com/library/ and http://www.pmboulevard.com/Library.aspx . Because there is no single correct taxonomy, the ideal ECM should enable consumers to dynamically generate multi-faceted library views such as “show me templates organized by sector then customer then project management lifecycle.”
Incidental information location happens when a user intentionally sets out to locate information and another path to the information presents itself. This type of information discovery is often overlooked, but is an essential function of an ECM system through related content, targeting, personalization and recommendations. When consuming content, users should be presented with related content which can also include files and other supplements to the main body of information. Targeting is the matching of content facets to user facets. One example of targeting might be Virginia-based employees seeing a message on the corporate intranet that North Carolina-based employees do not. Information consumers can also locate information incidentally through recommendations that include such things as “this week’s most popular articles,” and “people who were interested in this were also interested in this.”
Consumers may also find information by accident. This does not mean it was purely by chance, however. The user may not have been expecting to find information there, but though the ECM system, its content managers made a conscious decision in content placement. By highlighting content in a newsletter or a featured location on a web page, users may find themselves reading an article they currently have no interest in, but equips them for a future endeavor. The ECM system should facilitate accidental information location by enabling content managers to feature content and control its placement.
Key indicators found in the Consume phase are:
Support for intentional, incidental and accidental location of information
Ability to identify and locate subject matter experts
Basic and advanced search
Targeting and recommendations
Ability to feature and control content placement
Enable information consumers to provide feedback
The last phase of the ECM lifecycle is content disposition. All content has a lifespan of relevance, timelessness and usefulness and the ECM should archive content beyond its “sell by date.” During the Publish phase, content managers should be able to specify the publication end date, and in this case archival should be automatic. For other content, both objective and subjective decisions must be made to archive.
The ECM system should assist content managers in determining what to archive or destroy. The system should provide usage analysis to identify unused content. In addition, by enabling information consumers to provide feedback about its usefulness, content can either be updated or archived.
Key indicators found in the Archive phase are:
Provide usage statistics to recommend content for archival
Enable content managers to make archival decisions based on information consumer feedback
By leveraging ECM, organizations can facilitate the transformation of personal information to shared information and tacit knowledge to explicit knowledge. By “knowing what it knows,” an enterprise can use experience as a foundation, rather than continually re-doing and re-learning on each task. Hopefully this series on ECM has imparted the importance of managing content at your organization and that content and ultimately knowledge management is a continuous improvement process. By understanding the ECM lifecycle and the attributes of a healthy system, you should be inspired to set goals and undertake an ECM improvement process in your organization.