Relevancy tuning is partially experimental as projects have different use cases, datasets, and business objectives. Some features are best used as part of your global search design to broadly enhance relevancy, while others are intended for targeted search use cases that may require significant relevancy adjustments. The following table offers directional guidance on how specific features address problems with relevancy:
| Relevancy feature | Optimization for search relevancy | Intended tuning type |
|---|
| Search Fields | Optimize relevancy using data from indexed fields that are most pertinent to your use case | Global |
| Synonyms | Enhance relevancy when indexing does not match users' understanding of common keywords such as acronyms | Global |
| Stopwords | Remove mundane search results by focusing on significant terms and ignoring commonly used words | Global |
| Ranking | Improve relevancy based on trigger conditions related to indexed data, such as giving additional weight to specific data fields or values | Global |
| Data Filters* | Apply rules to include or exclude data from searches based on conditions related to indexed data | Global |
| Rules* | Adjust relevancy based on trigger conditions related to search or query patterns performed | Targeted |
| Promotions* | Arrange specific ordering of results based on trigger conditions related to the search performed | Targeted |
| Smart Match Assist ** | Provide AI recommendations to your synonyms based on semantic understanding of your data and historical searches | Targeted |
| Smart Ranking ** | Enhance relevancy based on user intent using AI analysis of search with re-ranking of returned results | Global |
Relevancy is not the sole method for guiding users to find what they seek. Enhancing the search experience with UX features that users can engage with helps them quickly refine or redirect their search if the initial relevancy is suboptimal. The following are Search UX features that aid users in intuitively finding what they need with less time and fewer searches or clicks.
| Search UX feature | How it helps users |
|---|
| Sorting | Enables users to organize search results based on search relevancy or the sorted order of content items for any specified field, such as numeric, string, boolean, or date. |
| Faceting | Enables users to narrow down content in a search by fields defined and indexed with your data, such as Content Type, Year Published, or Site Name. |
| Spell Check | Provides users with a set of results for a similarly spelled search term when the original search yields no results or when a misspelling is detected. |
| Auto-Suggest | Assists users in finding search suggestions quickly based on historical searches with clicks or dictionaries defined by your content team. |
| Related Searches ** | Helps users identify alternative searches related to their original search based on historical searches with clicks. |
| Popular Searches ** | Helps users identify searches that have been most popular in the last 30 days for the Search App. |
*Features are available in the Premier and Premier Plus plans.
**Features are only available in the Premier Plus plan.