Shopify redirect planning: method and QA checks

A migration can look finished and still break access to the pages that matter. The catalog is live and checkout works, but historic URLs return errors, an old product lands on an unrelated collection, or international visitors are sent to the wrong market. That usually isn't one missing row. It is the result of redirect planning that started too late, with an incomplete inventory and no clear approval process.

A redirect plan connects every useful legacy URL to the best available destination on Shopify. It is an SEO, product and engineering deliverable. It also protects journeys outside search: bookmarks, press coverage, paid campaigns, email, printed QR codes, comparison engines, partner feeds and links saved in internal tools.

Google recommends mapping old URLs to new ones, using permanent server-side redirects and avoiding redirect chains during a site move. Shopify provides URL redirect management and supports CSV import and export. Those features implement the plan. They do not decide where each URL should go.

What the redirect plan needs to contain

A working file needs more than an old URL and a new URL. It should explain the decision, identify who approved it and hold the QA result.

Keep the file usable by the team that will run the store. A spreadsheet driven by unexplained formulas and understood by one person is a launch risk.

1. Freeze the source inventory before mapping

Start during planning, well before go-live. No single source describes the complete legacy site.

• A crawl finds URLs reached through internal links and reports status codes, canonicals, depth and existing redirect chains.
• Google Search Console shows URLs known to Google and helps identify pages that still receive impressions or clicks.
• Analytics can reveal active landing pages, provided the tracking and reporting window are understood.
• CMS, PIM or commerce platform exports provide business objects, identifiers and publication status.
• Backlink reports, campaigns, feeds and partner records surface entries that may not appear in navigation.
• Business teams know about seasonal pages, old promotions, printed QR codes and links still used by customer service.

Keep the raw snapshot. Do not immediately discard duplicates, parameters or redirects already present on the old site. They can explain residual traffic or expose a chain that needs to be shortened. Build a normalized working set afterward: separate host and protocol, use consistent path casing, classify parameters and remove fragments when they never reach the server.

Record the collection date and coverage of each source. A three-month analytics export and a crawl run today describe different windows. The plan should not hide that difference.

2. Stabilize Shopify targets before writing rules

Redirect mapping depends on the target information architecture. Product, collection, page and article handles need to be stable enough to use as destinations. If teams continue changing them without a decision log, the redirect file becomes stale after every edit.

For each historic page, choose the destination that serves the same need. A product that is still sold should reach its new product page, not the homepage. A legacy category should reach the matching collection. An editorial guide should lead to content that actually covers the original topic.

A loosely related destination may remove a visible error while disappointing both visitors and search engines. Destination relevance matters more than the presence of a 301 alone.

3. Build rules that can be reproduced

A small store may allow line-by-line review. A large or international catalog needs rules plus a controlled exception list.

A repeatable rule can transform a URL family when the target identifier or handle is known. Test it against a varied sample, then compare the output with approved exceptions. Do not infer a match from the title alone. Two products may share a name, titles change, and translations can produce colliding slugs.

Use a stable source identifier to connect the legacy object to its Shopify equivalent whenever one exists. Record the matching method too: exact ID, approved business table, manual equivalence or consolidation rule. That provenance tells reviewers which rows carry the most uncertainty.

Before import, check at least the following:

• every source URL is unique;
• rows marked for redirect have a target;
• no loop eventually sends a URL back to itself;
• no known chain remains in the plan;
• every target exists in the release candidate;
• domain, language and market stay aligned;
• fields contain no stray spaces or malformed values;
• approved rules are separate from unresolved rows.

Shopify documents the CSV structure used for redirect imports. When possible, start from a redirect export generated by Shopify Admin. That reduces header and formatting mistakes. Keep the approved file dated and owned rather than passing unnamed copies between inboxes.

4. Test before the domain switch

Pre-launch QA cannot always reproduce the final public-domain behavior, but it can remove most avoidable defects.

Load the file in the intended environment and confirm Shopify Admin accepts every row. Export the redirects again, then compare the count and content with the approved source file. A successful import does not prove that targets exist or make sense.

Prepare an automated test that reads the mapping, requests each source after cutover and records the first status, hop count, final URL, final status and response time. Before launch, the same script can validate final destinations and find 404s, protected pages, unpublished targets or handles that changed after approval.

Add a manual review for priority cases:

• pages that received organic clicks;
• URLs with confirmed backlinks;
• top landing pages;
• important products and collections;
• brand and editorial content;
• separate markets, languages and domains;
• manually created exceptions.

The manual review answers the question a script cannot: does the final page still meet the intent of the source page?

5. Coordinate the cutover

Redirects are one part of launch. Their activation needs to line up with DNS, Shopify resource publication, canonicals, hreflang, robots rules, sitemaps, analytics and external feeds.

On launch day:

1. freeze handles and the approved mapping file;
2. confirm that all priority targets are published;
3. load or activate the redirect rules;
4. switch the domain according to the approved runbook;
5. run the automated legacy URL test;
6. fix loops, priority 404s, unpublished targets and market errors first;
7. crawl the new store through its own internal links;
8. verify sitemaps and Search Console properties, then monitor indexing.

Name the owners in advance. SEO owns relevance and prioritization. E-commerce confirms product and content equivalence. Engineering owns implementation and HTTP behavior. The launch lead makes the go or no-go decision when a blocker remains. In a BlackSwan project, those responsibilities are set during planning according to the contracted scope.

6. Monitor beyond launch day

Monitor legacy URL tests, internal crawls, Search Console, analytics landing pages and HTTP requests missing from the inventory. Correct menus, content, campaigns and feeds at their source when they still use old URLs. Redirects are a safety net for historic entry points, not the routing layer of the new store.

Frequently asked questions

Should every legacy URL be redirected?

No. Inventory all known URLs, then make a decision. A page with no equivalent should not automatically go to the homepage. Traffic, links, business purpose and the reason for removal determine the right treatment.

Is a redirect enough to protect SEO?

No. It preserves a path to the new resource. The migration also depends on content, internal links, canonicals, hreflang, indexability, sitemaps, performance and post-launch monitoring. Our Shopify SEO migration page covers that wider scope.

Can the mapping wait until client acceptance testing?

That is too late. Target URLs need to inform architecture decisions and must be testable before cutover. The Shopify launch QA checklist includes a dedicated gate for SEO and redirects.

Who approves the destination for an old product?

SEO should not decide product equivalence alone. E-commerce or merchandising confirms the business relationship, SEO reviews intent and risk, and engineering verifies the implementation.

Discuss your Shopify migration

We plan URL changes alongside data, integrations and launch QA. That gives the team approved decisions, named owners and a monitoring plan before the domain moves.

Discuss your Shopify migration