Content Quality Score

Content Quality Score

Every page Opensolr crawls gets a content quality score from 0 to 1: how much real content the page carries. A full article scores high. A listing page, a tag page or a template page that only repeats its own description scores low.

It matters because thin pages often match a search on their title alone and push the real answers down. The score lets you push them back.

What raises the score

• Text. How much text the page really has. The first few hundred characters count the most.
• More page than summary. The text should be clearly longer than the page's own description.
• Text, not just markup. A heavy page with little readable text looks like a template.
• A real title and description. They help a little, never more than that.
• Structured data on the page (JSON-LD) adds a small bonus.

Text is the base and everything else only adjusts it, so perfect tags cannot lift a page that says nothing. A page with no text at all scores 0.

Use it in your ranking

1. In the Control Panel, open your index and click WebCrawler.
2. Click Settings and open Search Tuning.
3. Move the Content Quality Boost slider from Off toward Maximum.

It is off until you move it. It changes the order of the results and never removes one: the further you move it, the further thin pages drop. Details: Content Quality Boost.

Pushed documents

Documents you send with Data Ingestion get no score. The boost counts them as average, halfway between a thin page and a rich one. The score is stored in the field quality_f: Fields Opensolr adds to each document.