Enterprise Search

Enterprise search is the practice of identifying and enabling specific content across the entire organization to be indexed, searched, and displayed to authorized users.

How does Enterprise Search affect your organization? When individuals within your company are searching for documents, are they finding the correct and most current information? When customers visit your website, are they locating the products and services with ease? These are essential questions to make sure your company maximizes its profitability.

Business intelligence and enterprise search solutions allow both your employees and customers to locate the information they need to make the most informed decisions. Universal Enterprise Search is the key way in which companies organize their information and allow individuals to locate exactly what they are looking for.

Companies may have few managed content repositories. It could be few content management systems, knowledgebase applications, wiki applications, CRM, ERP, etc. They are usually isolated from one another. Companies also may have unmanaged content repositories such as network drives. These repositories grow fast thus resulting in disconnected enterprise knowledge assets.

In a situation like this, it would be a good idea to have a search tool that would allow users to search all these repositories at the same time. "Universal Search" term was born. Another term that was born is "Content Management Interoperability Services" (CMIS) which offers to connect some of those repositories. Companies are looking into Enterprise Search (ES) as a shortcut to their findability problems.

Enterprise search is not an alternative to ECM systems. Even though there are search applications that allow the search across different platforms, enterprise search alone is not a solution to the findability problem. While enterprise search provides great value, it does not mean that you can give up integrating your repositories or leave your content on network drives.

Whit Andrews, Gartner VP, distinguished analyst, and author of Gartner's 2009 "Magic Quadrant for Information Access Technology" report (Gartner includes enterprise search as a part of "Information Access Technology") stated that with having right enterprise search you can get search that will be very effective, but it isn't the same as 100% success. He also acknowledged that you cannot simply buy, install, and walk away from powerful ES systems. Both ES and ECM systems require diligence and governance.

Leslie Owens, a Forrester Research analyst, is also skeptical about the likelihood of an ES quick fix, although she too is a strong proponent of targeted search applications.

In addition, out-of-the-box search may get you part of the way to findability, but [it] generally works best when enhanced with supporting semantic information-such as controlled vocabularies.

Every content repository should have two access points - search and browse. Users are going to use the Search access point if they know exactly what they are looking for. They are going to enter keywords in the search field and retrieve results. If they retrieve too many results, the presence of metadata will allow them to limit the search to specific results.

If users don't know what they are looking for, they are going to browse the content to get some ideas of what is available in the content repository. At some point during browsing, they may switch to search.

Therefore having only search as one access points is not enough to ensure findability of content. Metadata and taxonomy are invaluable tools to ensure that search and browse access points work properly. Taxonomy that is intuitive would greatly help the browse access point. The successful taxonomy is the one that is not too deep and not too wide. The users should be able to get to content with as few clicks as possible and yet the list of documents should not be too long. Users are not going to scroll through long lists of documents.

Naming conventions are extremely important. If users can figure out the contents of a document from its name without opening it, it will greatly speed up the time the user will spend browsing for documents.

Having information governance in place will make sure that the content resides in the correct place, that is has correct metadata, and that the content follows procedures to keep it accurate and up-to-date.

So is it ECM or ES? The answer is both.

In my future posts, I will review search applications and will further address search strategy. I will also review few case studies in ECM and ES.