Near duplicate content
What it means
These pages are not identical but are more than 91% similar, usually the same text with a different heading, price or date. Search engines treat them as competing versions of one page.
How to fix it
Decide whether the differences matter to a reader. If they do, make them substantial: different openings, different detail, different examples. If they do not, consolidate the pages as you would for exact duplicates. Thin pages generated per location or per variant are the usual source.
Finding it in CrawlStat
After a crawl, open Issues and pick Near duplicate content. Click Show all in Pages to filter the Pages table to the affected pages, or Export affected to hand the list to whoever owns the fix. Audit ID: near_duplicate_content.