Query Enhancement

Search surfaces recommended corrections when a user may have misspelled a search term — supported in multiple languages. Yext's custom-trained model predicts potential spelling errors and corrections while referencing the content in your CMS. This helps suggest corrections to brand-specific terms like professional or product names. Plus, experience training lets you provide feedback on spellcheck suggestions to constantly improve your experience.

Stemming improves the algorithm's ability to match queries to relevant CMS content by trimming words down to their more basic forms, or "stems." For example, the word "Otolaryngological" in the query "Otolaryngological care near" might be stemmed to "Otolaryngolog." This would help the algorithm match the term to a Content field for doctor's specialties, like "Otolaryngology." Yext's custom stemming algorithm is purpose-built for industries with particularly complex terminology, like healthcare and financial services.

Search can recognize when a query includes a synonym (or synonyms), and will take the synonyms into account when determining which results to return. You can define synonyms for your brand, including one-way synonyms (e.g., "square" can be used in place of "rectangle" but not vice-versa) and synonym sets (groups of phrases that can be used interchangeably, like "car" and "automobile").