Don't miss Data Summit, May 22-23. Learn about big data, AI, machine learning, cognitive computing, blockchain, and more.

Trends In Enterprise Search

This article is part of the Best Practices White Paper Search/Information Access [May 2009]

   Bookmark and Share

As a fast-growing, dynamic component of today’s information navigation and discovery landscape, enterprise search has moved from being a niche-focused capability to becoming a strategic component of a corporation’s IT infrastructure. As organizations gain a fuller appreciation for where this technology is heading, their expectations for solutions built on it will grow apace.

Here are five important trends that we see becoming critical aspects of future enterprise search based solutions:

1. Search fully integrates with the desktop. Some of the most timely and relevant content found within an organization lives at the edges of the network. It’s the content scattered on desktop drives and local file stores—or trapped in the notes and attachments found in emails. Though of incredibly high value, content like this is often poorly organized, difficult to share effectively and easily lost. There will be a greater emphasis on tools that can securely crawl individual user desktops and provide context and structure to everything stored there. These tools will be able to integrate this local content with other broader-based information sources users may have available to them, and provide them with a rationalized, unified, contextually relevant way to navigate through it. Beyond that, users will look for ways to share content by thematic or contextual classifications rather than at a simple folder level. Organizations will be increasingly interested in finding ways to tap into the underutilized potential of this local information pool.

2. Web search federates to the cloud. A large part of the information that people work with on a daily basis lives outside the enterprise. News, product information and competitive insights are all available from many respected sites across the Web or in documents published by clients, vendors, suppliers or industry experts.

There are easily tens of thousands of qualified sources publishing hundreds of thousands of updates a day. They’re written in a variety of languages, many requiring translation. Some updates come from feeds while others require complex crawling. Each update needs to be analyzed and tagged with any number of valid taxonomies. The task of integrating all of this content can be daunting, and the data management infrastructure required to store it all is significant.

Firms will be reluctant to undertake this infrastructure build-out themselves. Instead, the Web discovery component of enterprise search will be provided as an externally hosted, highly configurable cloud-based service that can easily participate in federated search architectures. This will become the most flexible, cost-effective way for firms to integrate the wealth of content living outside their firewalls.

3. Clustering becomes critical. As more and more content is brought under the enterprise search umbrella, having the ability to group related items together—clustering—will grow in importance. Clustering allows for the aggregation of duplicate and highly similar content elements. This is extremely important when dealing with content sets created in social architectures where syndication and re-publication are commonplace. Clustering will also be required for the automated grouping of related but distinct content elements. This is important for managing disparate information sets that evolve over time and need to be presented in aggregate to accurately reflect the knowledge contained on a topic. Without effective clustering, even well-filtered information sets can be challenging to scan through.

4. Discovery becomes social. Most people aren’t looking to simply consume information—they want to interact with it. They want to refine, augment, focus and apply it to the specific business opportunities they need to address. All of the value and understanding that gets associated with information during this process needs to find its way back into the search infrastructure to benefit the rest of the enterprise. This can include custom tags, comments, newly created derivative content or other content items that may have been bundled or associated with it. The value of enterprise search will be significantly enhanced by refining the context provided by automated tagging and classification technologies with the insight and understanding of the people who use the information associated with it.

5. Taxonomies become personal. Information has traditionally been organized using taxonomies provided by publishers or aggregators. Though useful, this approach has several limitations. It imposes a perspective on content that isn’t always aligned with the way the consumers of that information may think about it. It also can make it difficult for people to search and navigate across information coming from different unaligned sources. With improvements in semantic technologies, there will be increased demand from individuals or teams to superimpose their own personal taxonomies onto the content they need to work with. These personal taxonomies will become the glue that ties together content from their desktops, file shares, enterprise data stores and the Web.

What the future holds.
Enterprise search will continue to evolve in ways that let people add value to the content they discover, and share it with everyone else in an organization. It will become the catalyst for a virtuous cycle of information discovery, augmentation, and collaboration that ends up maximizing not just the value of stored information, but also the effectiveness of everyone in an organization who’s using it.

InfoNgen brings a unified, consistent and uniquely relevant view to all the information that powers your enterprise—whether it’s content living in various silos within your organization, including email and files on employees’ desktops, or content outside your firewall, including public Web and paid subscription sources. InfoNgen’s automated semantic analysis, industry-specific taxonomies and high-speed, globally federated search technology provide an efficient and cost-effective way to deploy enterprise-class discovery solutions.

Search KMWorld


Buyers' Guide
Learn More in the Buyers' Guide!