What Your Site Search Index Needs

Three things

Every site search feature runs on an Opensolr Index that has these three things. Check them once, before you fill the index.

01 · A region with site search

Not every Opensolr region runs site search. On Add New Index, the regions that do have an orange outline, Web Crawler after Shared or Dedicated, and the line AI Hybrid Vector Search at the bottom. Today they run Solr 9.00 to 10.00.

• An index in one of them has the WebCrawler and Chat Bot sections in its Control Panel.
• An index in another region cannot do site search. Create a new index in one of these regions and fill it there: creating an index.

02 · The Opensolr schema

Site search reads and writes the fields of the Opensolr schema: the title, description and text of each page, its address, its meaning vector, and the fields for PDF pages, language and prices.

• The Drupal module and the WordPress plugin install it on your index when you connect them.
• For any other index, one API call tells whether it has the schema and what differs, and another installs it: check and install the schema. Installing replaces the configuration of the index, so documents already in it must be sent again.
• If you replace the schema of the index with your own, the Chat Bot shows a warning that it may not answer well.

Every field, explained: schema reference.

03 · AI on your plan

Meaning vectors, picture meaning, search by image, AI Hints, AI Reader and the chat bot use the AI of your plan. Every plan includes a monthly AI allowance of 500 AI requests; the pricing page offers bigger allowances.

Everything else works on words alone: crawling and pushing your data, search by words, facets and Query Elevation.

• What counts as an AI request, and what happens when the month's allowance is used: what site search uses from your plan.
• What each plan includes: API Quota.