1. Search Administration Guide
  2. Global Search UX and Supported Use Cases

Global Search UX and Supported Use Cases

Global Search with Object Filter
  • Unified Global Search Bar: Placed at the top of the application screen, it offers an intuitive, user-friendly interface that is seamlessly integrated with OpenSearch and accessible throughout the application.
  • Object Filter: User clicks the 'Object' filter and uses the drop-down list to choose one or multiple search categories. The filter enables global search across multiple data objects in Siebel application, helping user refine search results for efficient data discovery and retrieval.

Results Page

When performing a search in Siebel application using Redwood theme, the search results are presented in a full page with 5 columns:

  • Column 1-4: Top 4 searchable fields (as defined in Search Category → Searchable Fields in Siebel application).
  • The first field marked as Searchable becomes the drillable field. Choose a required field to serve as the drillable field.
  • Column 5: Search category (as defined in Search Category in both Siebel application and YAML file)

When performing a search in Siebel application using Aurora theme, the search results are presented in a vertical panel located on the left-hand of the screen. Each result consists of the following information:

  • Top line: Search category (as defined in Search Category in both Siebel application and YAML file).
  • Top searchable fields (as defined in Search Category → Searchable Fields in Siebel application).
  • The first field marked as Searchable becomes the drillable field. Choose a required field to serve as the drillable field.

Keyword Search

Keyword search refers to searching for exact words or phrases within indexed documents. It matches user queries against specific terms in the data, returning results that contain those keywords without interpreting the context or meaning behind them. This type of search is precise and relies on exact matches rather than natural language understanding.

Query Examples:

  • Query “Abbey General Hospital” → 1 matching record returned.
  • Query “NHS Trust” → "Derbyshire Royal Infirmary NHS Trust", "Nottingham City Hospital NHS Trust", "Kings Mill Ctre Healthcare NHS Trust".

Semantic Search

Semantic search is a search technique that understands the intent and contextual meaning behind a query, enabling it to find relevant results even if they don’t contain the exact keywords used.

Query Examples:

  • Details on auto supply
  • Exception Handling SR's
  • All open service requests
  • Accounts assigned to CELLIS
  • Symptoms of System crash
  • Opportunities created by MNASH
  • Closed SR's
  • Contacts from "Metro Motors Michigan" organization

Ranking the Search Results

Siebel Intelligent Search performs a complex hybrid search that combines keyword-based and semantic search techniques. The results retrieved from the OpenSearch engine include a relevance score, which quantifies how closely each result matches the search query. Siebel displays these results in descending order of relevance - meaning that results with the highest match quality (and therefore the highest relevance scores) appear at the top.

The Siebel Search Adapter uses a normalization processor to scale each relevance score to a value between 0 and 1, and then ranks or sorts the results accordingly. It’s possible for multiple results to have the same normalized relevance score.

The number of search results displayed is determined by two parameters in the modernsearch configuration YAML file: maxResults and percentOfTopScore.

First, the result with the highest relevance score is identified. A threshold score is then calculated by applying the percentOfTopScore value to this top score. For example, if the highest relevance score is 0.9 and percentOfTopScore is set to 40, the threshold becomes 40% of 0.9, which is 0.36.

All retrieved results with a relevance score greater than or equal to 0.36 are considered. The total number of such results is then compared with the value of maxResults, and the smaller of the two is used to determine the final number of results shown to the end user.

In the above example, if 15 results meet the threshold of 0.36, and maxResults is set to 20, then 15 results will be displayed to the user.

Fuzzy Search

Fuzzy search is a search technique that finds matches even when the search terms have typos, misspellings, or slight variations. It helps retrieve relevant results when exact matches aren’t available by allowing some errors or differences in the query.

Examples:

  • Query “Hincklyt & Bosworth PCG” → Auto Suggest will still prompt for the correct entry: "Hinckley & Bosworth PCG"
  • Query “Abbeyz Cardiology” → Auto Suggest will still prompt for the correct entries: "Abbey General Cardiology", "Abbey General Cardio Research", "Abbey General Cardiac Care"

Fuzziness in search can be enabled by setting the fuzzyEnabled parameter to TRUE. This parameter is located in the search section under the downstream section of the modernsearchconfig.yaml file. By default, fuzziness for the search operation is set to FALSE. Enabling fuzziness may result in additional results being retrieved during a search.

For Auto Suggest (or Auto Complete), a separate fuzzyEnabled parameter exists under the suggester section (also within the downstream section), and it is set to TRUE by default. Therefore, if the default values are not changed, users will experience fuzziness during typing (i.e., in Auto Suggest), but not in the actual search results. Additionally, the fuzziness parameter in both the suggester and search sections is set to AUTO by default, which is considered the optimal setting for fuzzy matching.

File Attachment Search

Locates specific files attached to Siebel business objects by searching through file names, content, and metadata.

Example:

  1. Enable OpenSearch File Attachment Search in Siebel Application:
    1. Enable 'OpenSearch FileIndexing' system preference: set it to 'Y' in you want to include file attachment search.
    2. Set 'OpenSearch Index File Path' system preference: To Location to store decompressed files temporarily during indexing. Ensure the file path is accessible and has write permissions.
  2. Modify "Literature ModernSearch" search category if needed, generate index.
  3. Choose "Object" filter In Global Search Bar, select "Literature ModernSearch" category.
  4. Search for "FAQ". A list of documents is returned.

Supported File Formats for Siebel Intelligent Search:

File Type File Extensions
HTML .htm, .html, .xhtml
PDF .pdf
PPT .ppt, .pptx, .pptm
Word .doc, .docx, .docm
XLS .xls, .xlsx, .xlsm
RTF .rtf
Note: Files smaller than 5MB will be fully included in the search. For files exceeding 5MB, only metadata such as file name, author, and description will be searched and retrieved.

Auto Suggest

Auto Suggest provides real-time, predictive search query suggestions to users as they type, helping them complete their queries faster and more accurately by anticipating their intent. The default character limit for Auto Suggest is 200.

Note: Visibility is not applied while performing Auto Suggest.

Auto Suggest Within a Single Category:

  • If the selected category is of type Attachment, the suggestions will be generated from indexed fields such as ActivityFileExt, ActivityFileName, etc but not from attachment content.
  • If the selected category is a non-attachment type, suggestions will be generated from indexed fields including embeddingData.

Auto Suggest Across All/Multiple Categories:

  • If any selected category is of type Attachment, suggestions will not be generated for it.
  • Suggestions will still be generated for non-attachment categories.

Search History

Search history refers to a saved list of a user’s past search queries, enabling easy access and quick reuse of previous searches.

Search history is specific to user login. Each user can only see his/her own search history.

Clear Search History:

Choose 'Clear Search History' at the bottom of the drop-down menu to remove saved search queries. Note that search history cannot be cleared by re-logging into the application, restarting the Siebel Server, or clearing the browser cache. Clearing the search history using one login does not affect search histories saved under other logins.

Search Query Optimization

For optimal search accuracy and performance, follow these recommendations when constructing your search queries:

Tip 1: Enclose the category name in square brackets [ ].

Examples:

John in [Contacts ModernSearch]

John in [Contacts]

John in [Contacts ModernSearch, Accounts ModernSearch]

John in [Contacts, Accounts]

Tip 2: Enclose any specific information that needs to be searched in double quotes " ".

Examples:

"John" in [Contacts ModernSearch]

[SR] in “Open” status

[Oppty] about "medical information"

[SR] with "Contact" information

Note: If the you do not follow these tips when performing a search, the search will still function, but some irrelevant results may appear.
Note: The following operations on Child Business Component are not supported in Intelligent Search:

  • Incremental Indexing

  • Access Control

  • Record drill-down

Examples: Service Request Attachments, Account Attachments, Opportunity Notes, Product Key Features, etc.