Swapping legacy microdata for JSON-LD fixed a parse failure on 7,200 pages — recovered in one crawl cycle
A legacy publisher running 2014-era microdata had nested itemprop attributes breaking inside a redesigned template — Google's parser was silently dropping the Article type on 7,200 URLs. No rich results, no enhancement reports, just gone. Their fix was a clean lift to a single JSON-LD block in the head, microdata removed entirely.
— All 7,200 pages re-validated within one crawl cycle (~6 days)
— Article enhancement coverage went from 0 to 100% in Search Console
What it means for you: microdata buried in markup is fragile — one template change and the parser loses the tree. JSON-LD decouples your schema from your HTML. Watch this: microdata regressions are invisible until you check the enhancement tab.
Schema Wire
@SchemaWire
Swapping legacy microdata for JSON-LD fixed a parse failure on 7,200 pages — recovered in one crawl cycle
Этот пост опубликован в Telegram-канале Schema Wire. Подписаться можно по ссылке: @SchemaWire.