Moving hreflang from HTML head to XML sitemap on 120k URLs
Hypothesis: at scale, sitemap-based hreflang would be processed more completely than head tags, because bloated heads were being truncated. Methodology: one large marketplace, ~120,000 URLs, 8 markets. The head-tag version added 56 alternate links per page, ballooning the <head>. We migrated the entire annotation set into hreflang sitemaps. 14-week window.
Findings:
— Pre-migration, GSC reported hreflang annotations on only ~71% of the cluster; we suspect partial parsing of oversized heads, though we couldn't confirm truncation directly.
— Post-migration, annotation coverage in the International Targeting report rose to ~96% over 9 weeks.
— Correct-country serving improved most on deep, low-authority pages (the long tail), where head tags had apparently been dropped: tail-query correct-serving +18 points.
— Head terms barely moved — they were already covered.
Caveats: GSC retired the standalone International Targeting report's hreflang error section partway through monitoring at some properties, so our coverage metric relied on URL inspection sampling for the last weeks — a methodology change mid-study. The 71%-to-96% figure should be read with that caveat.
Conclusion: sitemap hreflang scales better specifically for large, deep clusters. For a small site, the maintenance overhead likely isn't worth it.
Hreflang Lab
@HreflangLab
Moving hreflang from HTML head to XML sitemap on 120k URLs
Этот пост опубликован в Telegram-канале Hreflang Lab. Подписаться можно по ссылке: @HreflangLab.