SharePoint 2010 : Customizing the Search Results and Search Center - Introducing SharePoint Search

1. Benefits of the Search Features

SharePoint 2010 is able to search and index both structured and unstructured data while bringing together data spread across different applications and server environments. This requires the SharePoint Search Server, along with the content administrators, to import data into the SharePoint Search system, define categories that define the data, and expose those relationships to the user. SharePoint 2010 Search platforms provide the following features.

• Delivers similar content quickly to users through a single interface

• Provides a rich query language to support varying levels of customizations

• Promotes frequently visited pages or content higher in the search results

• Customizes search using built-in, customizable Web Parts

• Allows integration with external content providers

Providing users with these search functionalities is traditionally cumbersome and difficult to manage, but the SharePoint 2010 Search technology provides straightforward development and administration interfaces that include (but are not limited to) the following considerations.

• Manually adding or updating content to pages is expensive and it is difficult to meet expectations with content.

• Custom development is generally expensive and requires custom knowledge of search applications. By utilizing reusable business logic with user interface (UI) components and existing or new SharePoint infrastructure, IT organizations can leverage the same development and administrative staff for both functions.

• Because SharePoint Search is part of SharePoint 2010, you may see lower maintenance costs as a result of the framework that is very similar across all SharePoint applications.

All versions of SharePoint have benefited from the inclusion of the Microsoft FAST Search Server. Primarily, the new SharePoint version enhancements are centered on the user search experience and streamlined administration. Enterprises can now provide users with targeted, customizable, related information with an easy-to-use, conversational search experience. Furthermore, the search results provide key data that will direct the user to related information where more information on the search topic can be found. Finally, customizations to search parameters and targeted search phrases allow content managers to present targeted results, thereby directing users or customers to data using fewer searches and/or clicks of the mouse.

2. Introducing SharePoint Search

Providing users with relevant, useful information is a critical element in improving user efficiency in the workplace and on Internet-facing website applications. Search applications are one of the key ways companies can provide users with targeted information depending on their profile, previous search patterns, and peer rankings. SharePoint Search Server and SharePoint FAST Search Server 2010 have capabilities that enhance a user’s search experience and can be customized to suit the needs of the Web application.

Traditional search engines provided static search results in a one-size-fits-all methodology. A user obtains the same set of results, without determination or discrimination of a user’s needs or profile. For example, in a traditional search scenario, if a user searches for “gardening,” he will get results that describe the definition of gardening, images relating to the topic, and many other gardening resources. However, a more relevant search experience might show gardening ideas, listing for local gardening clubs, and advice for gardening in his local climate zone. SharePoint 2010 is able to provide the enterprise with better, more relevant search experience by controlling the user experience, data input, and the ranking of searched content.

To get a better understanding of these concepts, it is important to compare how users perceive the data and how search applications interpret data. For example, in the gardening search described previously, a user might type local gardening, intending to find a local gardening club or organization, but all the search engine understands is “gardening” and so presents the results accordingly. To better comprehend the challenges surrounding search results, you must understand the underlying source data complexities. SharePoint data, from the search perspective, should be viewed from two perspectives: the input data and the data the search engine sees or understands. The input data, or the documents included in a SharePoint library or list, is defined by the document properties. These document properties help define what the searching user should come away with during a search operation, along with the complex data inside the document file itself. On the other hand, the search engine only understands data according to its ranking schema. The ranking schema includes a defined list of words, document properties, and ranking information; these are used for retrieval and ranking during the search process. To better illustrate this concept, Figure 1 shows source data from a Microsoft Word document with the ranking schema a search engine might have.

Another challenge search engines face is determining the user intent. In the gardening example, the user typed local gardening in the search engine because he was looking for results on local gardening clubs; however, another user could type the same phrase when looking for local gardening climate zone information. In this case, the search engine needs more information to provide results that better suit each user’s intent. A user searching for content may have to try a search several times before finding the appropriate results for the query.

If you want to customize the SharePoint Search experience, you need to understand the different aspects of search quality (relevance). One of the most often used components of search is the conversational user experience (ConversationUI). The conversational user interface consists of the elements that make up the search page itself—the search results, related search results, and visual best bets. Another aspect of search quality is the quality of the data collection itself. Simply put, search results are only as good as the data provided. If a user searches for gardening, but no documents relating to gardening exist, the search returns no results. Finally, the ranking and keyword management allows the content to be ranked according to predefined rules.

Table 1 shows the Search features available in SharePoint 2010.

Table 2 describes the features available with SharePoint FAST Server Search 2010.