Knowledge article search tuning
Key Takeaway
To put a specific published article at the top of Knowledge search results, link promoted search terms to it. Synonym groups only widen what matches, so a synonym group alone won't force an article to the top.
Knowledge article search tuning is how an administrator helps users find the right Salesforce Knowledge article faster. Salesforce ranks articles automatically. Admins can add tools on top of that ranking: promoted search terms, synonym groups, search highlights and snippets, topics, and extra case keywords. Each tool solves a different search problem, so the skill is matching the tool to the need.
How Knowledge search ranks articles by default
Knowledge search uses the same search algorithms as the rest of Salesforce. These algorithms include tokenization, lemmatization, and stopword lists.
When a search has no operator, the search engine picks one. Many searches use AND, so every term must match. If no article matches all the terms, the engine falls back to OR. With OR, articles that contain more of the search terms rank higher.
Other factors that raise an article in the default ranking include:
- How often a term appears in the article
- How often the article is viewed or attached to cases
- Article ownership and recent activity
- Search terms that sit close together
- A search term that matches a word in the article title
Exact keyword matches rank higher than matches on synonyms or lemmatized terms.