WordPress to Shopify redirects: WooCommerce urls into /products/ and /collections/

WooCommerce adds its own url patterns on top of WordPress’s. The product base and the category base are configurable under Settings, Permalinks, so check what the old store actually used before you trust the defaults (WooCommerce permalinks). On the Shopify side, the prefix comes from the template type and only the handle after it is yours to choose (Shopify theme templates).

sub-categories — WooCommerce nests categories, /product-category/men/shirts/. Shopify collections are flat, so the nested url becomes /collections/mens-shirts or similar. The slug is no longer the last segment alone, which is why these rows land in needs review more often than products.

two urls per product on shopify — Shopify also answers /collections/name/products/slug and marks /products/slug as canonical. Always redirect to the /products/ form; Silentfrog reads the canonical tag when it crawls the new store, so the matched target is already the canonical url.

Shopify’s help center lists the prefixes a redirect cannot start with — any url beginning with them, not only the bare path: /apps, /application, /cart, /carts, /orders, /services and /shop, plus fixed paths such as /products, /collections and /collections/all (Shopify url redirects). Two of those collide head-on with typical WordPress sites:

• /shop. WooCommerce’s shop page is /shop/ by default, and with the shop base or shop base with category product permalink every product url starts with it: /shop/hoodie-with-logo/, /shop/hoodies/hoodie-with-logo/.
• /services. A service business on WordPress very often has /services/ and pages below it.

A row with one of those on the left is refused or ignored by the import. There is no setting that changes this. If the old store had its products under /shop/, the realistic options are a redirect layer in front of Shopify (a cdn or proxy that answers before the store does), or accepting the loss for the urls without traffic. Decide before the domain moves; Search Console tells you how much traffic those urls still get.

Shopify redirects a url only when it would otherwise be a 404. If a product, page or collection exists at the Redirect from path, the store serves it and the redirect does nothing. In a WooCommerce migration that matters in two places: a migration app that created a page with the old slug, and a Shopify page whose handle happens to equal an old WordPress path. The import accepts such rows without complaint, so check them after go-live, not in the import dialog.

The same page also caps the count: 100,000 redirects per store, 20 million on Plus. A WooCommerce store would need a very large catalogue to get near it, even though every row is written out individually.

WordPress has a sitemap index at /wp-sitemap.xml on a default install and at /sitemap_index.xml with Yoast or Rank Math; it points at one file per post type, products included. Silentfrog reads robots.txt and follows the index on its own. Add the Search Console pages export to the url list: it holds the discontinued products and old posts that no sitemap lists any more, and those are often the urls with the backlinks (finding the urls nothing links to).

Variation urls with query strings, ?attribute_pa_color=blue, are treated as the product itself: the tool ignores query strings, and Shopify’s redirect matches the path.

The matcher needs the new store’s pages, and a Shopify store in development sits behind a storefront password that a crawler cannot pass. Switch the password off for the run, on the your-store.myshopify.com domain, while the old domain still points at WordPress. The crawl takes minutes; the password can go back on afterwards.

Products match well because migration apps usually keep the WooCommerce slug as the Shopify handle, so slug and title agree. Categories and pages are where the fuzzy matches and the needs-review rows collect: collection handles get renamed, and /about-us/ becomes /pages/about. Sort by confidence and read those.

posts — a post at /post-name/ or /2021/03/post-name/ becomes /blogs/news/post-name. Shopify’s default blog is called News; if you rename it or add a second one, the handle in the middle changes with it. The date segments have no counterpart, which makes no difference on Shopify: every row is written out anyway.

category and tag archives — WordPress’s post categories and tags have no successor unless the theme shows tag-filtered blog pages. Redirect the archives that get traffic to the blog index and let the rest go.

uploads — a pdf other sites link to gets its own row with the file’s new Shopify cdn url as a full-url target. Images are rarely worth it.

feeds and admin paths — /feed/, /wp-admin/, /wp-login.php and /wp-json/ need nothing. Letting them 404 is the point.

The WordPress side may carry a redirect table in Redirection, Yoast Premium or Rank Math. Put those source urls into the url list, so they are matched against the new store directly; importing them as they stand would chain the old target to the new one.

The Shopify side may already have redirects too: Shopify adds one whenever a handle is changed with the redirect box ticked, and some migration apps write their own. Export the list from the URL redirects page before importing, and use the import’s overwrite option when your new map should win over what the app wrote (importing redirects into Shopify).

Every WordPress url ends in a slash; Shopify’s do not. Silentfrog treats /product/hoodie/ and /product/hoodie as one page and writes the source without the slash. Shopify’s documentation does not say whether a redirect for the slashless path also catches the slashed form, so include both forms in the check after the import: request /product/hoodie/ as well as /product/hoodie, since the slashed one is what the backlinks contain.

• Export the WordPress plugin’s redirects and the Search Console pages report while the old site is up.
• Take the storefront password off, and start a run with the WordPress domain as the old site and the myshopify.com domain as the new one. Paste the plugin’s sources and the Search Console urls into the url list (starting a run).
• Review needs review from the top, re-map categories and pages that went to the wrong target, and verify the rows you accept.
• Pull out every row whose old path begins with /shop or /services (the bare page and everything below it); those cannot go into Shopify.
• Download the redirects file with shopify picked, which writes the Redirect from,Redirect to header and every directory rule as single rows, and import it under Content, Menus, URL redirects.
• After the domain moves, crawl the old url list against the live store: one 301, then 200. Compare the imported count to the file; the difference is reserved prefixes or paths that still load.

The first 50 pages of the old site are free, which covers a small shop or a test of one category. Medium at 46 € per relaunch covers up to 150 old pages; a store with a few hundred products needs Unlimited at 150 €. Only the old site’s pages count.

• Shopify's prefixes are fixed: /products/, /collections/, /pages/, /blogs/<blog>/. Practically every WordPress url changes.
• No wildcards: one row per product, category, page and post. The shopify export writes them out for you.
• /shop and /services cannot be redirected inside Shopify. A WooCommerce store on the shop base permalink needs a redirect layer in front of the store.
• A redirect only fires when nothing exists at the old path.
• Take the storefront password off for the crawl, and test the slashed form of each old url after go-live.