Enterprise search: the Holy Grail of KM?
By Andy Moore, KMWorld publisher
At October’s KMWorld & Intranets 2003 exposition, more than half the exhibiting companies identified themselves as “enterprise search” providers. Besides playing hell with our category listings, this overwhelming imbalance posed a huge question:
“What’s going on with search?”
"What many businesses fail to realize is that they use enterprise search every day within a range of business applications, including CRM, ERP, portal, BI and e-mail," Yankee Group's Rob Lancaster, collaboration and content management senior analyst.
We have heard it before: Knowledge workers spend 25% of their time looking for information, according to the brain trust firms such as IDC and Delphi Group. Faced with such a productivity drain, business managers are urging their IT departments to find the right search tools to connect employees with relevant information faster. But with search tools embedded into practically every application, and consumer tools such as Google performing as viable business search tools, the choice comes down to low-price+free vs. high-price+performance.
“It’s clear that a significant percentage of users receive free software, particularly in the search market,” according to Lancaster. “This is likely the result of integrated functionality, rather than vendor giveaways.”
Search engines are viewed as cost centers with little or no revenue-generating upside, so developing the value proposition is elusive. But increasingly—as the content management story seeps in—organizations are beginning to believe that content can be reused and regenerated into profit. It’s slowly becoming a matter of altering stubborn perceptions: Search is a justifiable maintenance cost that is required in order to leverage the value of existing content.
Additionally, the “opportunity cost” of misplacing data—a special kind of risk aversion—can become the overriding decision driver. It takes time and money to recreate or reacquire lost information. Companies see enterprise search engines as the means to avoid the costs associated with recovering lost data.
Perhaps best of all, enterprise (also called “federated”) search engines perform outside of existing applications, relieving those applications from overhead processing duties. The legacy systems currently in place can avoid the additional burden of crunching queries, for example, and avoid wasting cycles that are better used to accomplish the intended purpose. Search engines also benefit from employing simpler user interfaces and Web-style constructs, and thus enjoy greater user acceptance—a serious consideration when introducing new, costly technology. This is probably the single greatest benefit, in fact: Locating documents/content residing in separate systems and existing in multiple formats through a simple Web browser interface is the highest expression of what we’ve been calling “knowledge management” for all these years. This is where it finally, actually pays off.
Looking ahead
The market for enterprise search software is mostly untapped, reckons Yankee Group in its 2003 Enterprise Information Management survey. “Thirty-eight percent of businesses with less than $100 million in revenue have made no appreciable investment in enterprise search licenses,” according to the Yankee report, while “20% of businesses with $500 million or more in revenue have at least 10 content-management systems in operation, compared with 3% of businesses with less than $500 million in revenue.”
“We estimate that the market for enterprise search will reach $475 million by the end of 2003 and will be worth $760 million by 2007,” declares Eric Woods, research director of Ovum. “These predictions are based on a continuing demand for better information access in all organizations, the strong growth in the closely related content management, as well as portals markets and important innovations in search technology that will open up new opportunities in coming years.”
Ovum identifies six “areas of innovation” to consider when shopping for enterprise search:
1. Extended categorization and taxonomy support. The experience gained in building intranet and e-commerce sites is driving the development of more flexible technologies for the definition, management and use of corporate taxonomies
2. Search analytics, which refers to the collection, analysis and exploitation of information about the way search technologies are used in practice.
3. Expertise and collaboration. User profiling, in various forms, is becoming a common feature in enterprise search and information discovery solutions.
4. Bridging the gap from unstructured to structured. There may eventually be a single approach to all information management in the enterprise, and therefore a single search or query interface to all information, but this is not something Ovum expects to see soon.
5. Multilingual support. E-commerce is boundary-free by nature as far as the language of users is concerned.
6. Visualization. Examples of the more integrated use of advanced visualization include tracking patterns in information flows in real time, and the use of graphic tools for understanding complex relationships across information sources.
The real transformation in search and information discovery, says Woods, will come from the integration of these different innovations. The vision is of a flexible categorization of information and knowledge sources (structured and unstructured, explicit and tacit), easily visualized and navigable, and tailored to the specific search requirements of both the organization and specific users.