Search Tuning
The Search Tuning tab in Settings → Opensolr Search lets you fine-tune how search results are ranked. All settings use Solr's edismax query parser under the hood.
The defaults work well for most sites. Only change these if you want to tweak how results are ranked or if you are getting unexpected search behavior.
Field Weights
These control how much each part of a page contributes to its relevance score. Higher weight means matches in that field rank higher. The values map directly to Solr edismax qf and pf boost parameters.
- Title — weight for the page title. Default:
0.03 - Description — weight for meta descriptions. Default:
0.03 - URI — weight for the page URL. Default:
0.01 - Text — weight for the full body text. Default:
0.01 - Text (structured) — weight for the cleaner JSON-LD structured text (
text_tfield). Default:0.03
Phrase boost multipliers are applied automatically: pf at 0.8x, pf2 at 0.4x, and pf3 at 0.2x of the field weights. These reward results where search terms appear close together.
Semantic ↔ Lexical Balance
This slider controls the balance between two types of search:
More semantic (slide left)
Better for natural language questions ("how do I set up search?"). Results match by meaning, not just keywords. Available on vector-enabled plans.
More keyword (slide right)
Better for exact lookups ("WooCommerce REST API"). Results must contain the actual words typed. Works on every Opensolr plan.
The slider maps to the lexical_weight parameter (range 0.1 to 1.0, default 0.2). Lower values give more weight to semantic meaning; higher values favor keyword matches.
Freshness Window
Controls how much recent content is boosted in results. The value is in days (range 2 to 365, default 7). Content published within this window gets a relevance boost. Older content is not penalized — it just does not get the freshness bonus.
Default Search Mode
Controls the primary sort order for results:
- Relevant (default) — score is the primary sort, timestamp is secondary. Best matches first.
- Recent — timestamp is the primary sort, score is secondary. Newest content first.
Minimum Match
Controls how many of the search terms must appear in a result for it to be included. This maps to Solr's mm (minimum match) parameter.
Flexible
2<65% 4<50% 8<40% — show more results even if not all words match. Good for broad discovery.
Balanced
2<90% 5<75% 8<60% 12<50% — most words must match. Recommended default for most sites.
Strict
2<95% 5<90% 8<80% — almost all words must match. Fewer results but very precise.
Custom
Enter your own raw Solr mm syntax for full control over matching behavior.
Search Mode
Controls how keyword and semantic signals are combined in the query. These options apply to vector-enabled plans — lexical keyword search works independently on every plan.
- Union (default) — broadest. A result matches if keywords OR meaning matches.
- Keywords Required — keywords must match; semantic adds extra boost on top.
- Meaning Required — semantic match required; keywords add extra boost on top.
- Intersection — most restrictive. Both keywords AND meaning must match.
Vector Candidate Pool
A slider from 10 to 1,000 controlling how many vector candidates are considered before final ranking. This maps to the topK parameter in the KNN vector query. Default: 500.
Higher values consider more candidates (slower but more thorough). Lower values are faster but may miss some relevant results. Leave at 500 unless you notice speed issues or missing results. This setting applies to vector-enabled plans.
Quality Boost
A slider from 0 to 1.0 (default 0.3) that rewards content-rich pages. The boost uses the quality_f Solr field, which is calculated at indexing time based on:
- Title length and quality
- Description length
- Body text length
- Presence of an
og_image
Set to 0 to disable. Higher values give more advantage to pages with rich, complete content over thin or stub pages.
Reset All to Defaults
A "Reset All to Defaults" button at the bottom of the page resets every tuning parameter back to the plugin defaults. This is useful if you have experimented with settings and want to start fresh.
Tuning changes apply to the next search query — no re-indexing or re-crawling needed. Try adjusting one setting at a time and testing with real queries to see the effect.