This content is part of the Essential Guide: The pros and cons of SharePoint continue to divide users
Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Enhancing SharePoint 2013 search

While you can build search functionality into SharePoint 2013, consider key issues, such as user-friendliness of the interface and data privacy.

Although much of the discussion surrounding the big data trend has revolved around storage, data protection and...

analytics, these are far from the only IT considerations. Another important issue is users' ability to locate the data that they need in a timely manner. Of course, doing so usually involves performing a search.

Enterprise search isn't new, but the technology has evolved considerably over the past few years in response to the explosion of data growth. SharePoint, for example, has its own built-in search. This built-in search app works relatively well and Microsoft continues to improve it with each version of SharePoint. With SharePoint 2013 search, Microsoft has added features such as hover preview of search results, and the ability to group search results based on the result type. For example, it is possible to display a group of PowerPoint documents in response to queries that include the word presentation.

And while it's possible to build on top of SharePoint, enabling more fine-grained custom search functionality, it's important to think about key issues upfront, such as user-friendliness of the interface and data privacy. First, consumer search engines such as Google have set the standard for enterprise search simplicity. Second, users should be able to retrieve only those results for which they have the permissions to view.

Beyond SharePoint search

Although the SharePoint search app continues to improve, some organizations find the built-in search capabilities to be inadequate. Increasingly, organizations have chosen to build their own SharePoint search apps. There are also a number of third-party vendors that provide enterprise search applications.

Of course, this raises the question about which features and capabilities a modern enterprise search application should have. Obviously, every organization has different needs. Even so, there are several key capabilities that seem to be universal.

The first is that the search application needs to be intuitive. Users are accustomed to search engines such as Google and Bing. These search engines make it possible to enter a few keywords and quickly receive a list of results. Users expect the same level of simplicity in an enterprise search.

The simplicity of the search process involves a few factors. The first is the interface. Although most enterprise search applications allow for complex queries, the vast majority of users do not know how to perform queries. As such, the interface should be simple, although it might include a link to an advanced search.

Although it is important for an enterprise search application to have a simple interface, there is another, more important aspect to simplicity. Consider how an Internet search engine such as Google works. The user enters a search query and is presented with a list of results. The user is then free to click on any of the results to see if the result matches his needs.

In an enterprise environment, it is important for the enterprise search application to adhere to the established security permissions. A result should appear only if the user who performed the search has access to the result. Displaying results that the user can't access is frustrating. Such results could also potentially pose a security risk if a preview of the restricted data is displayed.

Another must-have capability is application awareness. In a large organization, many different types of data could appear within a search query. The results could include email messages to videos to webpages. The enterprise search application must differentiate between data types and display search results in a meaningful way.

There are several ways of accomplishing this. Google, for example, displays tabs at the top of the result page. These tabs allow the user to filter the results by data type. For example, there is a tab for images, video and so on.

Some enterprise search applications ask users upfront what type of data they want. This approach works well because it allows queries to be made against relevant metadata.

For example, I recently saw a custom search application that asks users whether they want to search for people, documents or videos. If a user selects the people option, the search engine displays optional fields that allow him to search based on fields like name, department or title. If a user chooses to search on documents, he is presented with a standard search text box but also a series of checkboxes to filter based on document type. For example, a user might opt to search on Microsoft Word documents, but not PowerPoint documents.

Enterprise search continues to evolve. Whether an organization builds a custom enterprise search application or purchases a third-party search product, the challenges remained largely the same. The primary challenges that administrators will face include indexing multiple locations and data types, and filtering results based on a user's permissions. Addressing these challenges almost always requires the use of some sort of identity federation service that allows a user's identity to be validated regardless of the type and location of data that is being queried.

Next Steps

Search-based apps are changing information management

Improving content accessibility in SharePoint 2013

Search technology that 'makes' data-driven decisions

Dig Deeper on Enterprise search platforms

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

What kind of enterprise search applications does your organization use?
We've used Apache Solr, Elasticsearch, and ApexKB at various times. Searching applications are constantly evolving, and we've made a point of regularly testing out new applications - as well as examining the updates to old ones - to see if they're worth switching to.

Currently, we're limiting it to an annual review, since making changes any more frequently than that makes us feel like we can't settle on a product - and worse, it wastes valuable training time.