Tech Trend

5 Must-Haves to Succeed at Enterprise Search and Select the Right AI-driven Enterprise Search Software

Many visionary companies have implemented enterprise search that is highly successful as measured by adoption, usage, percentage of time users find information (within minutes), and the ability to personalize the presentation to the user’s needs. All of these implementations include a set of core components.  Within these core components are five must-have items, which are detailed below.

AIIM identifies enterprise search as “The practice of identifying and enabling specific content across the enterprise to be indexed, searched, and displayed to authorized users”, which is a concise and complete way to describe an enterprise search system.  To this we would add that enterprise search needs to do the same for experts.

Content Awareness

What is it?

Content awareness (or “content collection”) is the process of providing users access to all of the content (documents, emails, discussions, etc.) they need, regardless of the systems they are stored in.  Enterprise search takes this aggregation of content and provides the user methods for accessing and retrieving the specific and relevant information they are looking for.  Sometimes content is pushed to the user based on their profile or preference, while other times it is delivered to the user through a search and refinement process similar to how users interact with popular sites like Amazon, Expedia, etc.   From a technical perspective:

  •  In the push model, a source system is integrated with the search engine in such a way that it connects to it and pushes new content directly to the engine. This model is used when real-time indexing is important.
  • In the pull model, the software gathers content from sources using a connector such as a web crawler or a database connector. The connector typically polls the source with certain intervals to look for new, updated, or deleted content.

Why must you have it (#1)?

The reason is that without this your enterprise search will be a failure.

The key point here is to index ALL authoritative sources of content that has the relevant information that users need. Indexing all information is not necessarily the best approach to implementing an intelligent enterprise search platform. If a user experiences getting the wrong information, or the user has to leave the enterprise search platform to access another system to find content, then the utilization of the enterprise search platform will be impacted and the true value of enterprise search will never be realized.  Regardless of the enterprise search platform being used, getting the right data in the index is key.

Content Processing and Analysis

What is it?

Content from different sources will have different Meta tags or may have no Meta tags at all. Remember, Meta tags are key to finding relevant information. If Meta tags are not consistent, then the search results will not meet the user’s expectations and often will provide too much information or information that is not relevant. Furthermore, the ability to provide refiners (or facets) to let users refine the results so they can get to the right information quickly will be negatively impacted. The content processing phase’s job is to process documents into a format the search engine understands, and to normalize that content in various ways to improve recall or precision. These may include stemming, lemmatization, synonym expansion, entity extraction, part of speech tagging, etc.

Why must you have it (#2)?

The reason is that without this your enterprise search will be a failure.

For an enterprise search system to be effective, a must have is the ability to account for the organization specific data needs and structure.  A great way to do this is through a centralized taxonomy that utilizes automated content analysis to identify and associate keywords, Meta tags, and concepts with content.

A Flexible Index

What is it?

After connecting the line of business and numerous content systems within an organization, the textual content of a document is stored in an index, often also referred to as the enterprise search engine.  This index is optimized for quick lookups without storing the full text of the document. The index may contain the dictionary of all unique words found across all content as well as information about ranking and term frequency.  The question to ask here is if the chosen enterprise search engine has the flexibility required to meet the changing needs of your organization.  Many organizations find that over time, the various search engine options innovate at different rates, and new capabilities from other engines may be necessary to meet new requirements.

Why must you have it (#3)?

The key point here is flexibility.  There are multiple enterprise search engines to choose from, and organizations often find themselves tied to their choice for years to come.  The cost to migrate or switch is high, as new capabilities or innovations become available in alternate options.  BA Insight can enable your organization to completely index all necessary data into multiple enterprise indexes like SharePointElasticSearch, and Recommind Decisiv, delivering the must have of flexibility.  You would not have to choose a single index, rather you could seamlessly leverage multiple indexes to solve the varied business problems being faced.

User Experience

What is it?

The user experience governs how users interact with the system and find the content they are looking for.  This is their first impression of the system and goes a long way in providing an experience that meets their ultimate needs.  Through the user experience users interact with the enterprise search system by entering queries or search terms and are often provided navigational options such as refinement, paging, and filtering, as well as information on how many items were found and how long their search query took.

While this component of the search system seems the simplest, it is in fact the most important.  For any enterprise search system to be truly effective and to deliver on the promise of increasing operational efficiency, it must provide users with an experience that is easy to navigate and puts valuable tools at their fingertips.

Why must you have it (#4)?

Without a clean and easy to understand user experience, users will not use or adopt the enterprise search platform.  Additionally, without applications that provide value and increase usability to users, adoption will also suffer. BA Insight provides a number of applications designed to bring this type of value directly to end users, from advanced refinement that allows users to quickly drill to the specific data they are interested in, to instant document previews that allow rapid validation of the quality of the search result.  Any enterprise search system must have a user experience that brings real value to users.

Search Results

What is it?

When a user executes a search, their “query” is processed and compared to the search index, and the system returns results (or “hits”) referencing source content that is deemed most relevant to what the user searched for.

Within this component is where the differences between enterprise search systems can be seen.  The order in which results are returned is governed by the internal relevancy system of the search platform.  Often times, these default relevancy systems fail to account for external influences on the potential results.  Does it account for who issues the query, and how that person is different than the last person who entered a similar query?  Does it factor in organizational taxonomy, synonyms, and terminology applicable only within the specific organization?  Does it account for popular content, or allow the user to specify advanced criteria to quickly hone in on the item being searched for?

Why must you have it (#5)?

Control and access to the relevancy configuration that powers the search results is key.  Business users need the ability to tweak and tailor the order of results to meet the needs of business users.  In order for an enterprise search system to succeed, this level of control is a must have.

The Future of Enterprise Search

Steven Nicolaou, Principal Consultant at Microsoft and a key proponent of Microsoft Enterprise Search, feels that the future of enterprise search “will bring shorter innovation cycles, continuous user experience improvements, deeper integration with first and third party applications and more ETL-like functionality to handle poor quality content.”

Source: 5 Must-Haves to Succeed at Enterprise Search and Select the Right AI-driven Enterprise Search Software

Leave a Reply

Your email address will not be published. Required fields are marked *