Dev.to WebDev πŸ›  Dev πŸ‘ 0 πŸ“– 2 min read

Checking multilingual indexing after a Cloudflare Workers migration

I work on GoSIM.ch, a travel eSIM shop. We recently moved from self-hosted Django to Cloudflare Workers and added Simplified and Traditional Chinese. Tonight I checked whether the translated pages were actually reaching

I work on GoSIM.ch, a travel eSIM shop. We recently moved from self-hosted Django to Cloudflare Workers and added Simplified and Traditional Chinese.

Tonight I checked whether the translated pages were actually reaching search engines. The files existed, the pages rendered, and the sitemap included them. Google and Bing still showed different stages of discovery.

Count the translations, then inspect the live pages

Both Chinese locales have the same 42 blog entries as English. Our live sitemap contains 558 Chinese URLs, split evenly between the two locales.

That is useful coverage evidence, but it doesn't prove the pages are indexed. I also opened sample homepages, guides and a destination page to check the language, canonical URL, language alternatives and links to the relevant plans. A translated JSON file alone wouldn't catch a broken link in the rendered page.

Ask Google about specific URLs

Search Console's sitemap report showed a successful fetch today with no errors. Its URL Inspection API gave a more useful answer for individual pages:

  • Both Chinese homepages were submitted and indexed, with matching Google-selected and declared canonical URLs.
  • Both Chinese student-discount guides were discovered but not indexed yet.

The request body is small:

{
  "inspectionUrl": "https://gosim.ch/zh-hans/blog/student-discount-esim/",
  "siteUrl": "sc-domain:gosim.ch",
  "languageCode": "en-US"
}

That goes to POST https://searchconsole.googleapis.com/v1/urlInspection/index:inspect with an authorized Search Console connection. It reads Google's recorded status. It doesn't request indexing or make Google crawl the page immediately.

We checked coverageState, lastCrawlTime, googleCanonical, userCanonical and the recognized sitemap. A successful sitemap fetch and an unindexed guide can both be true.

Check Bing separately

Bing's sitemap report still showed its last crawl as November 1, while Google's latest fetch was November 6. Bing also had no usable crawl record for the four Chinese URLs we sampled.

I resubmitted our existing sitemap through Bing Webmaster Tools. The API accepted the request. That is a submission result, not an indexing result; we'll check the crawl and URL records again after Bing has had time to process it.

Keep a small release checklist

For a multilingual content change, our checks now cover source-file parity, rendered language, canonical URLs, language alternatives, useful plan links, sitemap inclusion and a sample of actual search-engine records.

I keep those results separate. "Published", "in the sitemap", "submitted" and "indexed" describe different things. It makes the next action clearer: fix a real page defect, submit missing discovery information, or give an already-discovered page time to be crawled.

πŸ“° Read the original article on Dev.to WebDev

Originally published by Dev.to WebDev. Aggregated on AIWithGhost for educational purposes β€” full credit and traffic to the original publisher.