Hreflang Validator

This hreflang validator checks every hreflang tag across your site at once: missing return links, pages that don’t list themselves, invalid language and region codes, conflicting URLs, and alternates that redirect or point to a different canonical. Drop in your XML sitemaps or a crawl export, and you get each problem grouped by type with the pages it affects, a reason, and a fix. Then copy corrected tags for every set of pages, or download a fixed hreflang sitemap. Everything runs in your browser, so 20,000 pages take seconds and nothing is uploaded.

How to use the hreflang validator

• Add your hreflang data. Drop in XML sitemaps (.xml or .xml.gz), a crawl export with hreflang columns such as Screaming Frog’s hreflang export (.csv or .xlsx), or a table with page, hreflang, and alternate URL columns. You can also paste a sitemap, rows, or a page’s HTML.
• Add crawl data if you have it. A crawl export with status codes, indexability, and canonical URLs lets the hreflang validator check the pages each tag points to.
• Review the issues. Errors, warnings, and notices are counted at the top. Select an issue type to list every page it affects, then copy those URLs for your developer.
• Check the other tabs. Pages lists the issues for each URL, and Language coverage shows which versions each set has.
• Use the fixed tags. The Fixed tags tab gives the complete, corrected list for each set, and you can download every issue as CSV, a fixed sitemap, or tags per page.

What the hreflang checker looks for

Google’s guide to localized versions of your pages sets out the rules: each language version must list itself and every other version, the pages must point to each other, URLs must be fully qualified, and codes must be an ISO 639-1 language with an optional ISO 3166-1 Alpha 2 region. This hreflang checker tests each of those, plus the problems that come from real sites, such as old URLs, redirects, and canonicals. There are 19 checks in all:

Some hreflang errors are only visible across pages, which is why checking one page at a time misses them. A hreflang validator that reads the whole sitemap or crawl can see that the German page doesn’t point back to the English one, or that two pages disagree about which URL is the German version.

Hreflang return links

Missing hreflang return links are the most common hreflang error, and the most damaging, so they are the first thing the hreflang validator reports. Google says that if two pages don’t both point to each other, the tags are ignored, so one missing tag can switch off hreflang for that pair. They usually come from a page added in one language but not linked from the others, a CMS that builds tags from a stale list, or a translated URL that changed slug.

The hreflang validator also catches a subtler version: the return link exists, but uses a different version of the URL. If the English page at https://www.example.com/en/ lists https://www.example.com/de/ and the German page points back to http://www.example.com/en/, the URLs don’t match exactly, so the link isn’t confirmed. The detail column says what differs: the scheme, www, or the trailing slash.

Hreflang codes that look right but aren’t

An hreflang code is a language, optionally followed by a region. The language comes first, so en-GB works and GB alone doesn’t. The common mistakes:

• en-UK: the code for the United Kingdom is GB. Google says UK has no effect.
• de_DE: use a hyphen, not an underscore.
• es-419: Latin American Spanish is common in other systems, but Google only supports two-letter regions, so use es or a country such as es-MX.
• dk, jp, gb: country codes used as languages. Danish is da and Japanese is ja.
• se and uk: valid, but se is Northern Sami, not Swedish (sv), and uk is Ukrainian, not British English (en-GB).
• zh-Hant and zh-Hans are valid: a script code can follow the language.

For any code, check the list of ISO 639 language codes. The hreflang validator suggests a replacement for each invalid code and uses it in the fixed tags.

Add a crawl to check hreflang targets

Your sitemap can be perfectly reciprocal and still point at the wrong pages. If an alternate redirects, returns 404, is noindex, or has its canonical set to another URL, the tag points somewhere that can’t be shown in search. Google’s guide doesn’t cover canonicals, but the safe rule is to point hreflang only at pages that return 200, are indexable, and are their own canonical.

Add a crawl export with status codes, indexability, and canonical columns next to your sitemap, and the hreflang validator checks every target against it. It also warns when a page that carries hreflang can’t be indexed itself. To fix redirected URLs, the .htaccess Redirect Generator helps, and the fixed tags already swap in each target’s canonical URL.

Language coverage and fixed tags

The hreflang validator groups pages that link to each other into sets, such as one product in every language. Language coverage shows how many sets include each version, so you can see at a glance that x-default is missing from one set or that a language only covers part of the site.

The Fixed tags tab builds the complete list for each set: every version, including itself, plus x-default where the set has one. Conflicts are settled by what most pages say, invalid codes are replaced, redirected and noindex targets are left out, and where pages disagree about a URL’s code, the code the page gives itself wins. Paste them into the head of every page in the set, or download a fixed sitemap. To build hreflang from scratch, use the hreflang Tag Generator, and to rebuild a sitemap from a crawl, the XML Sitemap Generator.

Check hreflang tags on one page

To check hreflang tags on a single page, open the paste box and paste the page’s HTML. The hreflang validator reads the canonical tag to find the page’s own URL; if there isn’t one, enter the URL. One page can’t show whether the others link back, so those alternates are notices until you add their HTML or a sitemap. For return links across a site, a sitemap or crawl export is always the better input.

Privacy

The hreflang validator reads your sitemaps and crawl exports in your browser, and nothing is uploaded. Nothing is saved either, and Start over clears everything.