From Gainsight Support:
We sometimes get questions about how the search in the community platform works exactly, and what can be done to optimize it. As the search is weighing up many factors, it’s not always possible to determine exactly why a particular topic is or isn’t appearing in the search results. However, in this article, we will provide some context on how the search works as well as some best practices that can be applied to optimize search.
Searchable Attributes
When a user enters a query, the search engine evaluates multiple attributes across both community and federated (external) content to determine matching results. These include:
-
Title of the topic or article
-
First post (content)
-
Whether it’s an article
-
Whether it has the best answer (if it’s a question)
-
Replies
-
Tags
-
Category and Parent Category
-
Metadata from external sources such as Help Centers or Learning Platforms (via federated search connectors)
These attributes are used across all indexed content, whether it originates from within the community or from approved external platforms. Structured tagging and categorization further enhance searchability and filter usability.
Result Ranking
Once matching content is identified, a scoring model is used to rank the results shown to users. This ranking is based on a set of weighted factors, applied equally across all sources and content types. These include:
-
The number of search terms that match
-
The proximity of the terms to each other within the content
-
Content type indicators, such as whether it is an article or a solved question.
-
Date of last update
-
Engagement metrics include:
-
Number of likes
-
Number of replies
-
Number of views
-
Number of likes
-
-
Spelling similarity for minor typos
Search results are ordered by relevance rather than by source or content type. Community and federated content appear in a unified list, and the ranking model ensures that the most useful and timely content is prioritized. Each content type (e.g., article, question, discussion) is displayed with a distinct visual label.
Best Practices for Optimizing Search Results
Search behavior is automated and relies on content structure and metadata. While the underlying algorithm cannot be customized, the following practices help ensure that content is discoverable and performs well in search:
- Use clear, descriptive titles that accurately reflect the topic
- Incorporate common search terms and user-relevant keywords into titles
- Convert high-value posts into articles for better ranking
- Mark the best answers on solved questions to improve visibility
- Apply consistent and relevant tags and categories to all content
- Remove outdated or irrelevant content that no longer provides value
- Maintain structured formatting and consistent terminology across internal and external sources.
External content should follow the same optimization practices to ensure it is accurately indexed and appears in relevant searches. Regular moderation and content audits can help maintain search quality across the community.
