|
|
||||||||||||||||
|
||||||||||||||||
| ||||||||||||||||
|
|
||||||||||||||||
|
||||||||||||||||
|
By Dr. Bernard L. Palowitch, Jr., President and CEO, Iknow LLC
Rapid product and service innovation requires almost fanatical attention to marketplace trends and to the actions of competitors. Given the huge volume of information now available from countless electronic sources over the Internet, keeping up-to-date should seem extraordinarily easy. On the contrary, our work with Global 1000 clients has revealed that even with this abundance of information, many executives complain about the lack of real intelligence available for making good strategic and tactical decisions. Delving deeper, we have found that many market research analysts and competitive intelligence professionals do not have the internal business processes and software tools necessary to make use of this information.
Three Common Problems with Managing Enterprise Information
Our consulting work in knowledge and content management has uncovered three common problems regarding information availability and access.
Business Value to the Market Research and Competitive Intelligence Functions
In most companies, the responsibility for keeping a watchful eye on the external marketplace rests with the market research, business development, and/or competitive intelligence functions. Researchers and analysts in these areas are responsible for reading and interpreting news, research reports, trade and industry publications, competitor websites, corporate filings, competitor marketing and sales materials, and much, much more.
Most organizations can greatly improve the way they manage this knowledge, and can derive significant benefits in return. Improving information availability and access can increase productivity by allowing employees to more efficiently locate content on the corporate intranet, on company portals, and in various internal databases and repositories. Specific examples include CI analysts having better access to materials published both internally and externally, allowing them to more rapidly assemble, synthesize, and report on competitive intelligence activities; the information analysis and synthesis process is enhanced because previous reports, report templates, historical data, and other internal content can be located efficiently; and incoming queries are handled more quickly. The impact is truly significant when you consider that a recent survey conducted by Delphi Research of over 300 companies showed that 73% of the survey population spent 4 or more hours per week, and that 29% spent more than 8 hours per week (at least the equivalent of a full work day per week!), looking for information.
We believe that there is a compelling business case for addressing information silos, information overload, and the poor organization of information within companies. We have worked with CEOs and senior management teams on business justifications based on leveraging intellectual assets across the organization, increasing the efficiency and effectiveness of teams and business processes, and improving the quality and speed of making decisions.
In the remainder of this article, we want to briefly introduce and describe several knowledge and content management technologies that form the foundation for improving information access, information retrieval, and information categorization. These technologies, listed in Exhibit 1, improve internal productivity and effectiveness by automating many of the routine tasks, thereby freeing up the time of analysts and researchers so that they can focus on more value-adding activities. These technologies have been implemented across numerous industries; companies have achieved positive returns on investment; and the availability of commercial software products make system deployment straight forward.
The risks of missing key marketplace information can be enormous—to a product, a brand, or even the company! Investments in these technologies will move market research and CI departments beyond “finding the needle in a haystack” to “connecting the dots” among the various nuggets of business and competitive intelligence.
Content Management Systems Conquer Information Silos
After making sizable investments in IT infrastructure and in enterprise software solutions, companies have found that they now have silos of content throughout the organization. Content is generated by different departments and divisions and housed in different, business-specific areas on a corporate network or intranet. A major emphasis of IT spending today is to provide access to these individual silos by connecting them together with content management technologies to allow that information to be leveraged across the enterprise.
Content management systems conquer information silos. Broadly speaking, a content management system (CMS) facilitates the processes of capturing, storing, finding, and reusing an organization's intellectual assets. Specifically, these systems offer the following features. For content creation, a CMS offers desktop authoring, use of document and report templates, support for taxonomies and metadata, multiple language support, and support for authoring communities. CMSs can control hundreds of thousands of pages on hundreds of intranet and Internet websites with dozens of contributors, allowing each user to work on the same pages without conflict. Content management features include version control and rollback, rendition management, audit trail, and the assignment and use of defined roles and responsibilities for the various users and stakeholders. Content control features include automated workflow, document and repository-level security, business rules for content publication and authorization, syndication, synchronization, and information integrity. Many systems offer on-demand content formatting and publishing to multiple delivery channels, including the Web, print, wireless/handheld devices, and cell phones.
Content management systems are being implemented to support key marketing, marketing research, and CI business processes. Some of the leading content management software vendors include Documentum, FileNet, Hummingbird, IBM, Interwoven, Microsoft, Open Text, Stellent, and Vignette.
Search and Text Mining Technologies Overcome Information Overload
Once content is put into logical, organized repositories, then the ability to apply good technologies for information retrieval becomes possible. Search and text mining are two technologies that can address the issue of information overload.
Search
Search has recently emerged as a critical component in information delivery, due to the overwhelming volume and availability of electronic information. The primary technology delivered through most enterprise search vendors is keyword-based full-text search. People seeking content in unfamiliar domains enter basic keywords or phrases into full-text search tools. Search results are prioritized, in part, by the density of the keywords in the text.
Simple keyword search is the norm for public-facing Internet websites. However, researchers and analysts inside a company are well familiar with the company’s internal resources. They need highly-accurate results across a vast library of material, and therefore, require a more sophisticated search solution that leverages a deeper understanding of the domain.
Companies can improve knowledge worker productivity by providing more advanced search capabilities in their intranet and corporate portal environments. Natural language processing is one approach that allows the search functionality to deliver effective results regardless of how the user submits his query. Context-based search is another approach at improving the relevancy of search results.
Some of the leading search software vendors include Convera, Copernic, Google, Intelliseek, IBM, Inxight, ISYS, Lextek, Northern Light, Primus, Verity, and YourAmigo.
Text Mining
Text mining software automatically “reads” huge amounts of text in unstructured electronic documents, databases, and repositories. This software follows a set of specific instructions, or logic, to extract relevant information and present it to the user in any number of tabular, visual, and interactive formats. Text mining software can perform a wide variety of analytic functions, including automatic summarization, content filtering, data extraction, automatic database population, and metadata creation. Text mining, while still in its infancy, offers hope for conquering the growing overload of electronic information.
Business value from text mining arises from the software’s ability to navigate large unstructured information sources quickly and accurately. Application areas include market research, competitive intelligence, many types of due diligence, and news analysis.
Some of the leading text mining and text analysis software vendors include Attensity, Autonomy, ClearForest, Convera, Copernic, Inxight, Megaputer, Readware, SPSS/LexiQuest, TEMIS, Verity, and Vivísimo.
Taxonomies and Metadata Help Organize Enterprise Content
Most companies recognize early on that implementing a search solution needs to satisfy very diverse end-user needs. A comprehensive search solution—one that utilizes taxonomies and metadata to structure and categorize enterprise content—will drive greater employee productivity.
Taxonomies
Taxonomies are used to organize information into meaningful frameworks. A taxonomy refers to the logical organization of content into related subject areas. Taxonomies are flexible structures and can be developed to cover many different topics to any desired level of granularity.
Taxonomies can be constructed to provide multiple organizational perspectives into the content within an enterprise. For example, a product manager may search for content on a competitor by initially concentrating on a product name. A corporate communications director may look for that same content much differently; he may first focus on a marketing campaign or a promotion type, then search by geography or by timeframe, and then lastly, by product.
The primary benefit of using taxonomies is that they facilitate intelligent search and retrieval. Once a categorization structure is developed, information can then be populated into the defined groups or clusters. Taxonomies reduce information disorganization because the structure allows searchers to rapidly and intuitively navigate the entire body of content to find pockets of related, topic-specific information.
Metadata
Metadata is a second approach for organizing and categorizing information. Metadata is structured, descriptive information that identifies and describes the various attributes of a piece of content within an application or environment. For example, the metadata that describe a third-party research report include its title, author, catalogue number, publisher, language of publication, and publication date. The values associated for each of these fields are called metadata values.
Search tools incorporating a metadata framework can deliver more reliable and consistent search results than search algorithms based on keywords alone. Using the same example, if the title and author were unknown, the report could still be quickly identified by entering the known metadata values for the publisher and/or date.
The key to effective metadata management is the development and ongoing evolution of a controlled metadata vocabulary that accurately characterizes the business domain and that satisfies the search requirements of specific end-user groups. Creating taxonomies and developing optimal metadata is quite difficult, since our language has many ways of expressing the same or similar ideas, as well as organizing those ideas.
Business value from information categorization and classification arises from the broader access and reuse of existing knowledge assets. The ability to find, share, and reuse these knowledge assets quickly, when needed, adds value by shortening critical business process cycle times and by reducing costs, yielding faster, better, and more cost effective business decision making.
Commercial software products are available that offer extensive taxonomy and classification toolsets to help with creation, management and customization of taxonomies and metadata. Some of the leading taxonomy and metadata management software vendors include Attensity, Ascential, Autonomy, ClearForest, Convera, Mondeca, Recommind, SER Solutions, SPSS/LexiQuest, Stratify, TEMIS, Verity, and Wordmap. Standardized or industry-accepted taxonomies are also becoming commercially available.
Putting It All Together
Iknow LLC has done extensive work in deploying knowledge and content management technologies in the marketing, business development, and competitive intelligence areas. Exhibit 2 illustrates a high-level process schematic of how source content is analyzed and transformed into actionable competitive intelligence for a $3 billion per annum pharmaceutical product. All of the five KM technologies discussed above play a role in this solution.
The process involves three major steps.
Conclusion
As collections of content become larger and more varied and business requirements for finding content become more demanding, enterprises need more sophisticated information retrieval solutions for delivering relevant content to users. Keyword-based search alone is insufficient.
In this paper, we describe five technologies that are being used today to address information access, information retrieval, and information categorization issues. A content management system is the optimum environment to guarantee consistency of metadata across the enterprise and to ensure that all content delivered to an intranet, portal, or Web application is categorized for easy search and navigation. An optimal search and classification solution typically involves a combination of traditional search modes and navigable search results. Multiple technologies, including classification, linguistic analysis, and the use of taxonomies, are usually required.
We are not aware of any single software product that covers all of the topics discussed in this paper. Business solutions typically integrate several different software applications (e.g., content repository, portal, workflow automation) to effectively address today’s knowledge and content management challenges.
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
