The basic problem hasn’t changed much: Ever since organizations began creating and storing content digitally, the inability to locate needed information at a later date has often stymied business users.
IT vendors have tried to solve the problem with enterprise search
But things are changing. “Satisfaction with enterprise search is increasing as search tools have improved in the past two years,” IDC analyst Sue Feldman wrote in a report published last September. Meanwhile, consulting firm Gartner Inc. expects the enterprise search market to grow from an estimated $1.3 billion in annual revenue to nearly $2 billion within three years.
As Feldman indicated, advances in enterprise search technology have made it easier to deploy and use. For example, the development of enterprise search appliances gives user organizations the option of buying packages that bundle together search software with server and storage hardware.
The appliance model isn’t a new phenomenon; it also has been successfully applied to other types of applications, such as data warehouse appliances. The appeal it could have as an enterprise search appliance to the business side within an organization is twofold, according to Adriaan Bloem, a Netherlands-based analyst at content management consulting firm Real Story Group. First, he said, appliances typically can be deployed quickly, as the combination of hardware and software comes preconfigured and pretuned.
Enterprise search on someone else’s dime
But Bloem noted that because appliances include hardware, funding for purchases often can be secured from the IT department’s capital-expense budget or another infrastructure-related pool of money, as opposed to line-of-business budgets. In short, business units and departments might get the benefits of an enterprise search appliance without having to pay for it out of their own pockets, Bloem said.
Enterprise search tools are also benefiting from the inclusion of text analytics capabilities, said Seth Grimes, principal consultant at Alta Plana Corp. in Takoma Park, Md. Grimes noted that Web search and enterprise search are two very different beasts. Online search engines processing keyword searches rank websites or pages based on their popularity with Web users and the number of times they’ve been linked to from other sites, among other factors.
“You just can’t do that with enterprise search,” Grimes said. “It’s not the number of links that matter but rather some other measure of relevancy.”
In a law firm, for example, a lawyer might urgently need to find an old and little-used document concerning an important legal precedent. A conventional Web search engine likely wouldn’t give the document a high ranking because of its low access level, Grimes said.
But more and more, he added, enterprise search engines are embracing text analytics to help mine Word documents, PDFs and other forms of unstructured data in order to pinpoint the content that is most relevant to a particular search query. Working together, Grimes said, enterprise search and text analytics tools “can really understand the meaning of content, and that is improving the quality of enterprise search on enterprise information.”
Speaking the language of enterprise search tools
Another development that is working in favor of enterprise search is the increasing use of natural language processing (NLP) technology in search applications.
Twenty years ago, users basically had to know another language to get any value from enterprise search tools, said Sharon Flank, founder of DataStrategy Consulting LLC in Grand Rapids, Mich. Now, she added, “it’s no longer the case that someone needs to know Boolean [logic] or some arcane language in order to make the search engine work.”
That’s because many enterprise search vendors have embraced NLP in response to pressure from customers to make searching for and finding corporate information as simple as it is to run searches on the Web. NLP-based enterprise search, Flank explained, enables users to enter search terms and phrases in the same way they typically would speak the queries.
The move toward NLP is particularly important for business users who have neither the time to learn complex query-building techniques nor any interest in doing so, Flank noted. “Users don’t like to put in long, complicated search queries,” she said. “Nor do they want to go to training.”
Technical advances such as NLP don’t eliminate all the challenges and complexity of deploying enterprise search technology. But, Frank and other analysts said, they are helping put enterprise search on a path toward increased adoption and user satisfaction.
ABOUT THE AUTHOR
Jeff Kellyis a freelance writer.