Boost types reference
We are actively working to improve this documentation. The content you see here may be incomplete, subject to change, or may not fully reflect the current state of the feature. We appreciate your understanding as we continue to enhance our docs.
This reference describes each boost type, when to use it, and how it works.
Where a boost applies
Every boost, regardless of type, has two optional settings that control where it applies. Both default to applying everywhere, so existing configurations keep working unchanged.
Search surfaces
Limits the boost to the Main CMS Search (tool-header and fullscreen search), Selector & Typeahead Searches (object-selector popups, the content-selector quick typeahead, and the rich-text enhancement picker), or both. Leave empty to apply everywhere.
Use this when a boost helps the main search but pollutes typeahead results (or vice versa) — e.g. a strong recency boost that should not reorder editors' object pickers.
Applied to searches
Gates the boost by the content types a search targets — e.g. the type of the field an object selector is picking for:
- All Searches (default) — the boost participates in every search.
- Only Searches For Specific Types — participates only in searches restricted to the selected types (a selector for one of them, or a type-filtered search). Searches that span other types — including the main CMS search — skip the boost, so it can never affect content outside the selected types.
- All Searches Except For Specific Types — skipped only when a search targets nothing but the selected types (e.g. a Person picker for "All except Person"). The main CMS search still applies the boost; note that a recency boost there can still influence the excluded types' ranking, because recency boosts cannot discriminate per result.
Selected types may be abstract classes or interfaces — their concrete implementors match automatically.
Behavior change: type scoping is now enforced
Prior versions offered a Type selection on index-based boosts that only filtered the field dropdown and had no effect on search results. Type selections are now enforced:
- On text matching boosts, the (now multi-select) Types selection restricts which content the boost lifts.
- On Newest/Oldest boosts, per-result type restriction is not technically possible; the Types selection still only filters the field choices, and the field is cleared automatically if a type change makes it unavailable. Use Applied to searches to control where a recency boost participates.
If a saved boost had a Type set purely to narrow the dropdown, it will begin restricting results after upgrading — clear the Types field to restore global behavior.
Text matching boosts
These boosts prioritize results based on how well field values match the search query.
Exact match
Boosts results where the selected field's value exactly equals the entire search query.
Configuration:
- Types — (Optional) Restrict the boost to specific content types; also filters the available fields. Leave empty to boost matching content of any type
- Index — The text field to match against
- Weight — Boost strength (0-100)
How it works:
- The field value must match the complete query string exactly
- Case sensitivity depends on the underlying search engine configuration
- Whitespace in the query matters
Best for:
- SKU or product code searches
- Title fields where exact matches should rank highest
- ID or reference number lookups
Example: If the search query is "Breaking News Update" and an article's title is exactly "Breaking News Update", that article receives the boost.
Partial match
Boosts results where any of the individual search terms appear in the selected field.
Configuration:
- Types — (Optional) Restrict the boost to specific content types; also filters the available fields. Leave empty to boost matching content of any type
- Index — The text field to match against
- Weight — Boost strength (0-100)
How it works:
- Splits the search query into individual terms (by whitespace)
- Boosts results where any term matches within the field
- More matching terms generally means higher relevance
Best for:
- General content searches
- Fields with descriptive text (titles, summaries, body content)
- When partial matches should still surface relevant results
Example: If the search query is "climate change policy", articles containing "climate", "change", or "policy" in the selected field receive the boost.
Starts with match
Boosts results where the selected field's value begins with the search query or any of its terms.
Configuration:
- Types — (Optional) Restrict the boost to specific content types; also filters the available fields. Leave empty to boost matching content of any type
- Index — The text field to match against
- Weight — Boost strength (0-100)
How it works:
- Checks if the field value starts with the complete query string
- Also checks if the field starts with any individual query term (when multiple terms exist)
- Useful for autocomplete-style searching
Best for:
- Name or title searches where users type the beginning of a value
- Autocomplete scenarios
- Alphabetically-organized content
Example: Searching for "Intro" would boost articles with titles like "Introduction to...", "Introducing the new...", etc.
Content type boost
Content type
Boosts all results of a specific content type, regardless of the search query.
Configuration:
- Content Type — The type to prioritize
- Weight — Boost strength (0-100)
How it works:
- Applies a relevance boost to all content of the selected type
- Also includes subtypes of the selected type
- Query-independent—always applies when searching
Best for:
- Sites where certain content types are more important (e.g., Articles over Images)
- Ensuring primary content surfaces above supporting content
- Editorial workflows where specific types should be easily discoverable
Example: Setting a Content Type boost for "Article" with weight 50 means articles will generally rank higher than other content types in search results.
Date-based boosts
These boosts prioritize results based on date field values.
Newest
Boosts results with more recent dates in the selected field.
Configuration:
- Types — (Optional) Filter the available date fields by content type. Does not restrict which content the boost lifts — see below
- Index — The date field to sort by (only date-type fields appear)
- Weight — Boost strength (0-100)
How it works:
- Content with more recent dates receives higher relevance
- The boost is proportional—very recent content gets more boost than slightly older content
- Null dates or missing values receive no boost
- The boost applies to all content that has the selected date field, regardless of the Types selection — recency boosts cannot be restricted per content type. Use Applied to searches to keep the boost out of specific searches (e.g. certain object pickers)
Best for:
- News sites where freshness matters
- Event calendars (upcoming events)
- Any content where recency indicates relevance
Example: With Newest boost on publishDate, an article published yesterday ranks higher than one published last month, all else being equal.
Oldest
Boosts results with older dates in the selected field.
Configuration:
- Types — (Optional) Filter the available date fields by content type. Does not restrict which content the boost lifts — see below
- Index — The date field to sort by (only date-type fields appear)
- Weight — Boost strength (0-100)
How it works:
- Content with older dates receives higher relevance
- The boost is proportional—older content gets more boost
- Null dates or missing values receive no boost
- The boost applies to all content that has the selected date field, regardless of the Types selection — age boosts cannot be restricted per content type. Use Applied to searches to keep the boost out of specific searches
Best for:
- Historical archives
- Reference content that ages well
- Situations where original/foundational content should surface first
Example: For a historical photo archive, setting Oldest boost on originalDate surfaces the oldest photographs first.
Semantic boost
Semantic match
Boosts results when search terms are phonetically similar to configured keywords, using Metaphone encoding.
Configuration:
- Content Type — The type to boost when phonetic matches occur
- Keywords — Terms that trigger the boost when phonetically matched
- Weight — Boost strength (0-100)
How it works:
- Converts search terms and keywords to phonetic codes using the Metaphone algorithm
- When a search term's phonetic code matches a keyword's code, the boost applies
- Also matches against the content type's display name
Best for:
- Handling common misspellings
- Multilingual sites where users might spell terms differently
- Synonyms or alternative terms for content categories
Examples:
- "nooz" phonetically matches "news"
- "artikle" matches "article"
- "foto" matches "photo"
Add common misspellings and alternative spellings of your content type names as keywords to improve search findability.
Choosing the right boost
| Use Case | Recommended Boost(s) |
|---|---|
| Exact ID/code lookup | Exact Match (high weight) |
| General content search | Partial Match + Content Type |
| News/blog site | Partial Match + Newest |
| E-commerce product search | Exact Match (SKU) + Starts With (name) |
| Historical archive | Partial Match + Oldest |
| Handling misspellings | Semantic Match |
| Prioritizing primary content | Content Type |
Weight guidelines by boost type
| Boost Type | Typical Weight Range | Notes |
|---|---|---|
| Exact Match | 60-90 | High because exact matches are strong signals |
| Partial Match | 30-50 | Moderate; shouldn't overpower exact matches |
| Starts With | 40-60 | Between exact and partial |
| Content Type | 20-40 | Subtle influence across all results |
| Newest/Oldest | 20-50 | Balance with text matching boosts |
| Semantic Match | 30-50 | Similar to partial match |