Shopify Hydrogen SEO checklist: SSR, JSON-LD, sitemap, robots

Live HTML and SSR

Start with the production response, not the component tree. Confirm the route returns HTTP 200, fetch the HTML, and compare it with what shoppers see after the page hydrates.

Important product titles, product descriptions, buying context, collection copy, product links, breadcrumbs, and pagination paths should not exist only inside client-only widgets when they carry search value.

• Product descriptions and buying guidance appear in initial HTML when they matter for the route.
• Collection pages expose crawlable headings, copy, product links, filters that do not trap crawlers, and pagination or load-more paths with a clear fallback.
• Tabs, personalization, reviews, recommendations, and app widgets enhance the page without becoming the only source of primary commerce content.

Metadata, canonical URLs, and route state

• Every indexable route has a route-specific title and meta description that match the page's actual intent.
• Canonical URLs use the final production domain and intentional handling for variants, filters, market paths, and campaign parameters.
• Do not assume Hydrogen will choose the correct canonical for every custom route. Treat canonical behavior as route logic you own and inspect.
• Noindex and robots meta directives are present only where they are intentional, documented, and checked on production.

JSON-LD

Google's structured data guidance rewards accuracy over volume. Add schema only when the visible page supports the properties you publish.

• Product structured data matches visible title, price, availability, variant, and image state.
• Article and guide pages expose Article schema.
• Breadcrumb schema matches the visible navigation path.
• FAQ schema is used only for visible FAQ content.
• Organization and Person facts stay consistent with approved public proof.

Sitemap and robots

Shopify documents Hydrogen sitemap and robots route patterns, but that does not remove implementation responsibility. Treat sitemap.xml, robots.txt, cache policy, canonical route membership, and Oxygen production behavior as outputs to inspect.

• Sitemap includes only public canonical URLs that should be discoverable.
• Robots behavior is checked on the final production domain, not only in a preview environment.
• If the storefront runs on Oxygen, robots.txt is verified on the public production environment with the custom domain attached.
• Old Liquid or legacy headless routes redirect cleanly before the sitemap points crawlers at the new URLs.
• Google treats sitemaps as discovery support, not a guarantee that every URL will be crawled or indexed.

Practical Hydrogen SEO checklist

1. Confirm each priority route returns HTTP 200 on the final production URL.
2. Fetch production HTML and verify important product, collection, and guide content is present before client-side interaction.
3. Inspect title, meta description, canonical URL, robots meta directives, Open Graph tags, and route-specific metadata.
4. Validate JSON-LD against the visible page state: product, breadcrumb, article, FAQ, Organization, and Person data should not invent facts.
5. Open sitemap.xml and confirm public canonical URLs are included while preview, filtered, duplicate, and noindex routes are excluded.
6. Open robots.txt on production and confirm crawl policy matches the launch plan.
7. Test redirects from old Liquid or legacy headless URLs before removing the old route map from the migration tracker.
8. Check internal links to the Hydrogen SEO service page, the SEO migration checklist template, the product SSR note, and the contact path.
9. Confirm llms-full lists the route as this site's AI-readable discovery support, without treating llms.txt as an official ranking factor.
10. Monitor Search Console after launch for crawl errors, canonical selection, indexing status, sitemap processing, and affected query/page changes.

Source verification and claim safety

Keep official facts separate from operator judgment. It is fair to say Shopify documents Hydrogen SEO setup surfaces for metadata, sitemap.xml, and robots.txt. It is not fair to claim Hydrogen automatically solves canonical strategy, route migrations, product schema accuracy, or indexing.

Use the official docs as the source boundary, then verify the actual storefront output. A claim is safer when it points to observable route behavior: a fetched HTML response, a canonical tag, a JSON-LD block, a sitemap entry, a robots rule, a redirect status, or a Search Console signal.

Structured data must match the visible page

• Product schema should match visible title, image, price, currency, availability, variant, and brand state.
• Breadcrumb schema should match the visible route hierarchy.
• Article, FAQ, Organization, and Person schema should be used only where the visible content supports it.
• Generated schema should not invent reviews, ratings, prices, offers, or business facts.

Rendered HTML is the first checkpoint

Product titles, product descriptions, buying guidance, collection copy, breadcrumbs, product links, and pagination should be present in the HTML when they carry search value. Client-only enhancements are fine, but they should not be the only place important commerce content exists.

This is where many Hydrogen SEO issues become visible. A polished PDP can still be weak if crawlers receive thin HTML and most useful content appears only after hydration.

Migration-specific SEO checks

1. Map old URLs to new URLs and test redirects on production.
2. Confirm canonical URLs use the final production domain.
3. Verify sitemap includes only public canonical URLs.
4. Check robots.txt after launch, not only preview.
5. Confirm internal links point at the new canonical routes.
6. Compare Search Console coverage, canonical selection, crawl errors, and indexed URLs after cutover.

AI-readable discovery without ranking promises

This site uses llms.txt and llms-full.txt as AI-readable discovery files so systems that choose to read them can understand the site's public pages and scope. Treat that as a site convention, not an official Shopify or Google standard and not a guaranteed ranking factor.

The useful overlap with AI discoverability is still practical SEO quality: clear headings, self-contained answer blocks, source-backed claims, accurate schema, working sitemaps, and internal links that describe the real relationship between pages.

What to hand off after an audit

• Route URL, status code, canonical URL, indexability, and sitemap membership.
• HTML evidence for missing or present product, collection, guide, breadcrumb, and pagination content.
• Structured-data findings with the visible field that supports or contradicts each property.
• Redirect findings for old Liquid, app, campaign, and legacy headless URLs.
• Search Console follow-up owner, timing, and the specific signals to watch after launch.