Magento to Shopify SEO Migration

WRITTEN BY
Kev Wiles
I’m a Fractional SEO Specialist with 12+ years’ experience working with eCommerce brands. I focus on making SEO simple, clear, and effective helping businesses cut through the noise and unlock real growth.
Moving from Magento to Shopify changes the address of every page on your store. Products, categories, CMS pages and blog posts all get new URLs, and Google has to learn the new ones and connect them to the old ones.
This guide covers my own process on how I have handled SEO migrations for brands moving from Magento to Shopify. It assumes the domain stays the same and only the URLs change.
What a migration does to your organic traffic
Expect some movement in organic traffic after launch. The size depends on how many pages you have and how much of your traffic comes from non-brand searches, since people searching for your brand will find you either way.
Shopify’s SEO lead, Kyle Risley, says any URL change has some impact on SEO traffic, with recovery taking a few weeks for small sites and a few months for large ones (Shopify).
His worst-case formula is a useful way to size the risk. Take the share of traffic that comes from organic search, multiply it by the share of that traffic that is non-brand, then multiply by 15% to 30%. The result is the revenue decline to expect over one to six months. His example is a store with 20% organic traffic, 70% of it non-brand, which gives a dip of 2.1% to 4.2% that recovers in one to three months. He adds that most migrations see declines of 10% or less (Shopify). Run your own numbers before quoting anyone a figure.
Magento stores carry four extra risks: a url_rewrite table that has grown for years, filter pages that rank on their own, a deep category tree that has to be flattened, and several store views serving different markets. Each is covered below.
What changes at platform level
These are the differences that matter for a migration, not a general platform comparison.
Area | Magento | Shopify | SEO impact |
|---|---|---|---|
Product URLs | Often at the root with a .html suffix, sometimes repeated under every category path | /products/handle is canonical, though the product also loads under a collection path | Several old URLs per product redirect to one new URL |
Category URLs | Nested paths such as /men/shoes.html | Flat /collections/handle | Hierarchy moves into navigation and breadcrumbs |
Category depth | Unlimited nesting | No native nesting | Deep trees are flattened and parent and child links rebuilt |
Filtered URLs | Layered navigation parameters, sometimes SEO-friendly paths from an extension | Filter parameters from Search & Discovery or a filter app | Filter pages that rank today do not exist unless built as collections |
Redirects | url_rewrite table, with server or CDN rules on top | Exact path, one URL to one URL, only on URLs that return a 404 | The map must be explicit and complete |
Robots.txt | A file or admin setting you control | Generated, edited through a robots.txt.liquid template | Rules are rebuilt on purpose, not copied |
Sitemap | Generated by Magento and configurable | An index at /sitemap.xml with sub-sitemaps, generated automatically | Submit the new index and retire the old one |
Server logs | Available from your host | Not available | Capture log data from Magento before launch |
Multi-store | Store views, usually with hreflang from an extension | Shopify Markets or separate stores | Store view prefixes and parameters need redirect rules |
URL structure and handles
Shopify uses fixed prefixes: /products/, /collections/, /pages/ and /blogs/ followed by the blog handle (Shopify). A Magento URL like /blue-shirt.html cannot be kept as it is.
You can control the handle, which is the last part of the URL. Matrixify, the import tool most Magento migrations use, builds handles from the Magento url_key, so a product keeps its slug even though the path around it changes (Matrixify). Identical slugs make the redirect map far easier to build and check.
Watch for collisions. Magento allows /mens/shoes.html and /womens/shoes.html because the category path makes them unique. Shopify handles must be unique within each resource type, so both cannot become /collections/shoes. Find these before import and rename them yourself, for example mens-shoes and womens-shoes, rather than letting a tool append a number.
Categories, collections and variant limits
Shopify collections are flat. A Magento tree of Men, Shoes and Running Shoes becomes three collections, and the relationship between them is rebuilt through the main menu, breadcrumbs, links on the collection pages and product assignment. Plan the linking at the same time as the collection list.
Variant limits have changed, and many guides still quote the old ones. Shopify raised the limit to 2,048 variants per product on 15 October 2025, with a maximum of 3 options (Shopify changelog). A Magento configurable product with four attributes, such as size, colour, material and length, still cannot become four options.
Robots.txt, sitemap and canonicals
Shopify generates robots.txt and sitemap.xml automatically. To change robots.txt you add a robots.txt.liquid template, which replaces the generated output. Shopify provides Liquid objects so you can keep its default rules and add to them (Shopify developer docs). Do not copy Magento’s robots.txt across, because most of its rules block Magento system paths that do not exist on Shopify.
The sitemap is an index pointing to separate sitemaps for products, collections, pages and blog posts, and you cannot edit it directly. The seo.hidden metafield can hide a product or page from it.
Product pages canonicalise to the clean /products/ URL. Filtered collection URLs are less clear. One practitioner source says they canonicalise to the clean collection URL and that filter parameters are not blocked in the default robots.txt, while merchants on the Shopify Community report filter combinations appearing in Google. I could not find Shopify documentation settling it, so open a filtered URL on your own store, view the source and read the live /robots.txt before deciding what to do.
Build the baseline
Before changing anything, I record what Google indexes, what earns traffic and revenue, what earns links, and what Magento does behind the scenes. Save every export with the date on it. This is the only way to prove what the migration did, and it decides which URLs get manual checking.
Segment the URL list by organic clicks, revenue and referring domains. The top slice is checked by hand on staging and again after launch. Shopify’s Magento guide suggests starting the redirect map with the top 500 to 1,000 URLs by organic traffic plus your existing redirects (Shopify). That works as a minimum, not a limit.
The Magento audit
A crawl only sees what is linked today. Magento remembers every URL a product or category has ever had, and a crawler cannot reach that history. I pull five things from Magento itself.
1. The url_rewrite table. It holds a row for every URL the store has served. The key columns are request_path (the URL a visitor requests), target_path (where it resolves), redirect_type (0 for an internal route, 301 or 302 for a redirect), entity_type and entity_id (product, category or CMS page), and store_id. A product renamed three times can have four request paths, and only the current one is in your sitemap. Adobe documents the redirect options for custom rewrites as No, Temporary (302) and Permanent (301), and notes that GET parameters are removed during redirection (Adobe Commerce).
As far as I can see, Magento 2 has no standard admin export for this table, so a developer runs a read-only query against a replica or copy. Add your table prefix if your install uses one.
SELECT url_rewrite_id, entity_type, entity_id, request_path,
target_path, redirect_type, store_id, is_autogenerated
FROM url_rewrite
ORDER BY store_id, entity_type, request_path;
Magento adds a redirect every time a URL key changes, which creates chains. This query finds redirects whose destination is the start of another redirect.
SELECT a.request_path AS first_url,
a.target_path AS second_url,
b.target_path AS third_url,
a.store_id
FROM url_rewrite a
JOIN url_rewrite b
ON b.request_path = a.target_path AND b.store_id = a.store_id
WHERE a.redirect_type IN (301, 302)
AND b.redirect_type IN (301, 302);
Export every store view, not just the default. Rows with redirect_type 0 are internal routing, not redirects.
2. The SEO configuration. In the admin this is Stores, Configuration, Catalog, Catalog, Search Engine Optimization. Record the product and category URL suffix, whether product URLs include the category path, whether Magento creates a permanent redirect when a URL key changes, and whether canonical tags are on. These settings explain why a product has the number of URLs it does.
3. The module list. Run bin/magento module:status or read app/etc/config.php and mark every module that touches SEO: canonical and noindex rules, schema, sitemaps, layered navigation, blog, brand pages, hreflang, reviews and feed exports. Each one is a function Shopify has to replace.
4. Internal search data. In the admin, Reports, Marketing, Search Terms shows what customers type into your site search, and it can be exported (Adobe Commerce). Terms with many searches and few results point to missing collections and content.
5. The CDN layer (Adobe Commerce Cloud only). Redirects may live in Fastly configuration instead of url_rewrite. Ask your host or developer whether any exist outside the database.
Crawl, Search Console, analytics, backlinks and logs
Crawl the full Magento site with Screaming Frog or Sitebulb and keep the export: status codes, titles, meta descriptions, H1s, canonicals, structured data, internal link counts and image alt text. Many Magento titles and descriptions are generated from templates and are not stored on the product record, so the crawl is the reliable record of what Google sees.
Export the last 16 months of Search Console data by page and query. Add organic landing page revenue from GA4 and a backlink export from Ahrefs or Semrush. A URL with no traffic today but strong links still needs a redirect, because the redirect carries the link value across.
Collect 30 to 90 days of server logs if your host keeps them. On Magento they often show Googlebot requesting old URLs and image paths under /media/catalog/product/ that are in no sitemap. Shopify gives no log access, so this is your only chance. After launch you rely on the Crawl Stats report in Search Console.
If the Magento front end is a PWA or headless build, such as Hyvä or Vue Storefront, crawl the rendered pages and confirm that the titles, canonicals and structured data in the browser match the crawl export.
URL mapping and redirects
Every indexed Magento URL needs a destination. I build the map from the url_rewrite export, the crawl, Search Console and the backlink data together, never from the sitemap alone. The most common failure is redirecting anything without an obvious match to the homepage or one generic collection. Google treats that as a soft 404 and the traffic gets nothing useful.
Magento URL patterns and where they go
Use this table as a starting list. Extensions change Magento’s URL behaviour, so confirm each pattern against your own crawl and url_rewrite export.
Pattern | Example | Shopify destination | Note |
|---|---|---|---|
Product with suffix | /blue-shirt.html | /products/blue-shirt | Handle usually matches the url_key |
Product under a category path | /men/shirts/blue-shirt.html | /products/blue-shirt | Many old URLs, one new one |
Category with suffix | /men/shirts.html | /collections/shirts | Check for handle collisions |
System product path | /catalog/product/view/id/123 | /products/handle | Resolve the ID through url_rewrite or the catalogue export |
System category path | /catalog/category/view/id/45 | /collections/handle | Same approach |
Filtered category | /shirts.html?color=49&price=10-20 | A collection if it earns traffic, otherwise the parent collection | Decide per combination |
Extension filter path | /shirts/blue | A purpose-built collection | Often ranks, check Search Console |
Sorting and paging parameters | ?product_list_order=price&product_list_limit=36&p=2 | The parent collection | Query strings may not redirect as expected on Shopify, test them |
Store view parameters and prefixes | ?___store=uk, /uk/ | The matching Shopify Market URL | Map per store view |
Internal search | /catalogsearch/result/?q=shirt | /search?q=shirt or a relevant collection | Only where it earned links or traffic |
CMS pages | /about-us | /pages/about-us | Includes any .html versions |
Blog and brand extension URLs | /blog/post-title, /brands/acme | /blogs/news/post-title, a vendor collection | Depends on the extension |
Media | /media/catalog/product/a/b/image.jpg | The image’s new CDN URL, set during product import | Images are not redirected, so image search traffic is re-earned |
Magento 1 legacy | /index.php/blue-shirt.html | /products/blue-shirt | Only if old links or logs still show them |
Mapping rules
Each old URL lands on the closest relevant page in one hop. A discontinued product goes to the closest collection or a true replacement, not the homepage. Flatten chains before import, so that A to B to C becomes A to C and B to C.
I work in tiers. The top URLs by traffic, revenue and links are mapped and checked by hand. The long tail is mapped by pattern, for example every product row to /products/ plus its url_key, then spot-checked. Pages with no traffic, no links and no internal links can return a 404. Shopify’s own guide makes the same point: a good 404 page with search can beat thousands of low-value redirects (Shopify).
Keep redirects live for a long time. Google’s Gary Illyes said in 2021 that the answer is at least one year (Search Engine Roundtable). Shopify says at least a year and ideally indefinitely (Shopify). Treat them as permanent, since old links never stop arriving.
Update internal links too. Links in CMS blocks, menus and blog posts that point to old URLs still work through the redirect, but each adds a hop.
Shopify redirect constraints and bulk import
The Shopify Help Center sets out the restrictions (Shopify Help Center). Each one affects a Magento map.
Redirects only work on URLs that return a 404. If the old path still loads a valid Shopify page, the redirect will not fire. Each redirect covers one specific URL, and the Help Center does not describe wildcard or pattern matching, so a Magento pattern rule becomes one row per URL. Query strings might not work as expected. Redirects cannot be created for tag-filtered collection URLs or from the reserved prefixes /apps, /application, /cart, /carts, /orders, /services and /shop. A .html URL cannot be redirected to the same URL without the extension, so map every .html path to an explicit /products/, /collections/ or /pages/ destination.
Standard plans allow 100,000 redirects. Shopify Plus allows 20,000,000. A Magento store with years of history and several store views can exceed the standard limit, in which case prioritise by organic revenue and links, as Shopify’s guide advises (Shopify).
To bulk import, go to Content, Menus, View URL Redirects, then Import. Upload a CSV and confirm the preview. The usual format is two columns headed Redirect from and Redirect to, with paths starting with a slash, and the Help Center links to a sample file. Import before pointing DNS at Shopify so the redirects are live when traffic arrives. Shopify also warns that 301 redirects can stay cached in browsers and search engines after you delete them, so test before you import.
What migration tools carry over
Matrixify takes a Magento product export and builds Shopify products from it. It converts categories to smart collections driven by tags, maps custom fields to metafields, creates handles from the url_key, and generates a Redirects sheet for old product URLs, old collection URLs and each collection and product combination. It supports a dry run (Matrixify). Cart2Cart and LitExtension are the other tools you will see recommended.
Treat tool-generated redirects as a starting point and reconcile them against the url_rewrite export, because a tool working from a product export cannot see historical URLs. Do not assume CMS pages, blog posts, custom redirects, SEO titles or meta descriptions come across. Test each on a sample before the full import. Shopify’s guide says the same of customer passwords, configurable products, custom attributes and CMS blocks (Shopify).
The redirect map workbook
I keep one master spreadsheet that every other step reads from. Columns: old URL, store ID, Magento entity type and ID, original target path and redirect type, the source that surfaced the URL (url_rewrite, crawl, Search Console, backlinks or logs), organic clicks over 12 months, referring domains, revenue, tier, new Shopify URL, how it was mapped, status code on staging, final destination status, number of hops, and who checked it. Sorting by tier and test status shows exactly what is mapped, tested and unresolved.
Faceted navigation, categories and landing pages
Layered navigation is the part of a Magento migration most likely to cost traffic quietly. Magento can generate an indexable URL for every filter combination, and many stores run an extension that turns filters into keyword-bearing paths such as /shoes/black. Some of those pages rank on their own. Shopify’s filtering produces parameter URLs, so a legacy filter landing page returns a 404 unless you rebuild it as a collection.
Deciding which filters become collections
Export 12 months of Search Console pages, isolate the ones with filter parameters or the extension’s path pattern, and rank them by clicks and revenue. Most will have no traffic.
For the ones that do, check that real demand sits behind them using Search Console queries and your internal search terms. Build a collection for each one worth keeping, either a smart collection with rules on tag, type, vendor or metafield, or a manual one, with its own title, description and copy. Redirect the old filter URL to it. Combinations that earned nothing redirect to the parent collection, or return a 404 if they never had links or traffic.
Agree the cut-off with the client instead of applying a universal threshold. A filter page with 20 clicks a month can matter more on a small store than one with 200 on a large store.
If you use a filter app such as Search & Discovery, Searchanise or Boost, check what it does to crawl paths and canonicals after installing it.
Category copy, CMS blocks and brand pages
Magento category pages often carry their SEO copy in the category description or a static CMS block, and neither appears in a standard product export. Check each priority category on the live site, copy what is there, and make sure the Shopify collection template outputs it as crawlable HTML, not inside a JavaScript tab or accordion. Shopify’s guide lists layout and content placement changes, including accordions, among the common ways a migration hurts SEO (Shopify).
Brand pages built by an extension become collections filtered by vendor. Anchor categories that pulled products from their children become smart collections with rules for those products. Where the Magento tree is more than three or four levels deep, Shopify’s guide advises flattening it and merging low-traffic categories into their parents (Shopify). Point the redirects from merged categories at the parent.
On-page elements, structured data and templates
I migrate like for like first and optimise afterwards. Changing platform, URLs and copy at the same time makes any traffic movement impossible to diagnose. Shopify’s guide lists unnecessary changes as the first common pitfall, including layout changes, new plug-ins and changed title tags (Shopify).
Metadata and content parity
Capture the rendered title, meta description, H1, body copy and image alt text for each priority page from the live Magento site, using the crawl export. Compare them with what is stored in Magento, since template-generated values often differ. On Shopify, the title and meta description are in the search engine listing section of each product, collection and page, the handle is the URL, and alt text is set per image. An unset SEO title falls back to a default that is often less relevant, which Shopify’s guide lists as a pitfall (Shopify).
Schema and reviews
Rebuild Product, Offer, AggregateRating, BreadcrumbList and Organization markup to match what Magento output, then validate it with Google’s Rich Results Test. Shopify themes typically ship basic Product JSON-LD, and review and SEO apps often add their own, so check each page for duplicate or conflicting markup. Review stars only survive if the review data is migrated to the new app (Judge.me, Yotpo, Okendo or similar) and the markup is output once. Export reviews from Magento as a backup first.
Duplicate product URLs through collections
A Shopify product loads at /products/handle and at /collections/handle/products/handle, with the canonical pointing to the clean product URL. Many themes link through the collection path using the Liquid filter product.url | within: collection, so internal links point at the non-canonical version. Link to product.url instead. Expect a higher “Alternate page with proper canonical tag” count in Search Console (Amsive).
Multi-store, hreflang and Shopify Markets
Each Magento store view needs a decision: a Shopify Market on the same domain, or a separate Shopify store. Markets on subfolders generally keep authority in one place, if inventory, pricing and branding allow. Separate stores suit cases where those differ too much for one catalogue.
Either way, redirect every store view prefix and parameter, such as /uk/ or ?___store=, to the matching Market URL, and map localised URL keys one by one, since a translated slug is a different path. Redirects apply across all market subfolders, though you can add separate redirects per subfolder when they need different destinations (Shopify Help Center).
Rebuild hreflang on purpose. Magento usually handles it through an extension. On Shopify it comes from Markets and the translation setup or a translation app, so check the live output. Confirm the primary market and default language too.
Magento extensions and what replaces them
Every module marked in the audit does something Shopify either does by default, replaces with an app, or cannot do. Vendors such as Amasty, Mirasvit, Mageworx, Aheadworks, Mageplaza and Magefan are examples of what to look for in your module list, not recommendations.
Extension type | What it did | Shopify replacement | Risk |
|---|---|---|---|
SEO suite | Canonicals, noindex rules, schema, sitemaps, meta templates | Theme and Shopify defaults, plus an SEO app only where there is a gap | Medium: apps can duplicate schema |
Layered navigation | Filters, sometimes with SEO-friendly URLs | Search & Discovery or a filter app, plus collections for filters that rank | High: ranking filter pages disappear |
Brand pages | A page per brand | Vendor collections | Low |
Blog | Posts under /blog/ | Shopify blog under /blogs/ | Medium: posts need redirects and metadata |
Hreflang | Alternate language tags | Markets or a translation app | Medium: verify output |
Reviews | Product reviews and rating markup | Judge.me, Yotpo, Okendo or similar | Medium: review history and stars |
Feed and Google Shopping | Product feed export | Google & YouTube app or a feed tool | High: product IDs can change |
Redirect manager | Bulk and pattern redirects | Shopify redirects, one URL at a time | High: patterns need expanding |
Do this audit before choosing the theme, because the theme and apps share some of these jobs.
Product types, attributes and variants
Magento product types do not map one to one onto Shopify’s product model, and this is where the set of indexable pages can change. Simple products that Magento marks as not visible individually, usually children of a configurable product, must not become standalone pages. Grouped and bundle products have no direct equivalent and need a bundles app or a rethink of how they are sold. Virtual and downloadable products need a decision on delivery.
Attributes become options, metafields or tags depending on use, and the team decides which remain as filters. Matrixify maps custom fields to metafields, which preserves the data but does not decide how it appears on the storefront (Matrixify). Shopify’s guide advises mapping custom attributes to metafields before migration and reviewing configurable products manually (Shopify).
Plan variants against the limits of 2,048 per product and 3 options (Shopify changelog). Each variant becomes its own item in a product feed, so a product with hundreds of variants creates hundreds of feed items.
Merchant Center and feed continuity
If you run Google Shopping through a Magento feed extension, your product IDs will change when the feed moves to Shopify. Google ties performance history to the product ID, so a new ID is a new product and the history does not follow.
Stores already on Shopify’s Google & YouTube app have a second issue. Channable reports that the app changed the IDs it sends to Merchant Center during the move from the Content API for Shopping to the Merchant API, from a country-coded format such as shopify_NL_123456_789012 to shopify_ZZ_123456_789012. The Content API for Shopping ended on 18 August 2026, and history tied to the old IDs cannot be restored once they are removed (Channable). Google also said it was deploying a fix for a separate bug that made some merchants lose product history after uninstalling and reinstalling the app (PPC News Feed). The two issues are different.
Before launch, decide whether to preserve your Magento item IDs through a feed tool or accept new IDs and plan for a learning period on Shopping and Performance Max. Keep GTINs, SKUs and brand values consistent, test the feed on staging, and keep structured data and feed attributes aligned so organic listings and the page agree on price and availability.
Staging QA, cutover and launch
Set go and no-go criteria before QA starts, with a rollback plan for anything that fails. A written list removes most of the arguments in the final week.
Timing matters. Shopify’s guide advises migrating at least six months before peak season, or right after it (Shopify). For a retailer with a Q4 peak, that means launching in the new year.
Staging QA
Block indexing on staging. Crawl it and compare against the baseline crawl: status codes, titles, H1s, canonicals, structured data and internal links. Test every redirect against the full list of old URLs, not a sample, using Screaming Frog in list mode with redirects followed so you see the final destination and status for each. Check Core Web Vitals on the new templates, and audit the third-party scripts that apps add, since those are the most common cause of a slower store.
DNS, proxies and the launch checklist
Lower the DNS TTL 24 to 48 hours before the change, confirm the SSL certificate has provisioned, and stage all redirects before pointing the domain, as Shopify’s guide recommends (Shopify). Check third-party DNS records such as MX and TXT first, because moving the domain is also how email breaks.
Be careful with proxies. Shopify merchants report certificate and connection problems when a proxy such as Cloudflare sits in front of a store (Shopify Community). Remove any WAF, CDN or edge rules that held redirects on Magento.
Check | Pass condition |
|---|---|
Password page and noindex | Removed, and the live page source has no noindex |
Robots.txt | Reads as intended at /robots.txt |
Redirects | Every priority URL returns a 301 to a 200 in one hop |
Canonicals | Self-referencing on products, collections and pages |
Structured data | Valid, no duplicates |
Hreflang | Present and reciprocal, if used |
Sitemap | /sitemap.xml loads and the old Magento sitemap is retired |
Tracking | GA4, GTM, conversion events and consent banner firing |
Search Console | Domain property verified, new sitemap submitted |
Checkout and forms | Test orders and contact forms work |
Rebuild every tag and event you had on Magento before launch, including GA4, Google Tag Manager, ad platform pixels and Bing Webmaster Tools. Many migrations lose a week of conversion data because tracking was left until last.
Post-launch monitoring
I check daily for the first two weeks and weekly for 8 to 12 weeks. Shopify’s guide recommends checking for 404s daily, running a full crawl within 24 hours of launch and cross-referencing it with the pre-migration URL list, and watching Search Console daily for crawl errors and index drops. It calls a 5% to 10% organic dip in week one common and a sustained drop of more than 20% a reason to investigate redirects and canonicals immediately (Shopify). These are vendor thresholds, so use them as a prompt to look, not as proof.
Period | What to check |
|---|---|
Days 1 to 3 | 404s, redirect errors, crawl status, indexing of key pages, tracking |
Days 4 to 14 | Indexed page counts, canonical selection, organic landing pages, conversion tracking |
Weeks 3 to 4 | Non-brand clicks and impressions by page type, rankings for priority queries |
Weeks 5 to 8 | Collection and product page performance against the baseline |
Weeks 9 to 12 | Recovery in non-brand clicks and organic revenue, backlink status, redirect gaps |
Separate brand from non-brand. Brand queries recover almost immediately and can make the total look healthy while non-brand is still down. Compare against the same weekdays and the same period last year to remove seasonality.
Reading Search Console
Indexed page counts usually fall after migration because Magento’s parameter and filter URLs were inflating the number. That is expected. “Page with redirect” and “Alternate page with proper canonical tag” counts will rise. Look for the loss of pages that mattered, not a lower total.
Average position can improve while clicks fall, because the long-tail pages that dragged the average down have gone. That is why success is measured on non-brand clicks and organic revenue. Shopify gives no logs, so use the Crawl Stats report to see how Googlebot treats the new store.
A non-brand decline that continues past a couple of months points to something structural: missing redirects, filter pages never rebuilt, lost category copy, or slower templates. Work back through the baseline to find which page type is losing.
After launch, update the URLs you control outside the site: paid campaign final URLs, email templates, social profiles, affiliate and marketplace listings and Google Business Profile. Then ask the top linking domains from your backlink export to update their links where that is easy.
Keeping AI search visibility through the move
People now ask ChatGPT, Google’s AI features and Perplexity the same questions they used to type into Google, and those systems cite pages. A URL change can break that too, because the cited page is now a redirect.
Before launch, run a fixed set of prompts about your category and record which of your pages are cited. After launch, run the same prompts and compare. The pages that carry your category knowledge, size guides, comparisons and policies are the ones to protect most carefully, with direct redirects and unchanged copy.
Migration timeline
Compress this for a small catalogue and extend it for a large one with several store views.
When | Task | Owner |
|---|---|---|
12 weeks before | Baseline exports: crawl, Search Console, analytics, backlinks, logs | SEO |
10 weeks before | url_rewrite export and extension audit | Developer and SEO |
8 weeks before | Filter page and collection decisions | SEO and merchandising |
6 weeks before | Redirect map version 1 and staging build | SEO and developer |
4 weeks before | Staging crawl QA against the baseline | SEO |
2 weeks before | Redirect import test, content freeze | SEO and developer |
1 week before | Go or no-go review, DNS TTL lowered | Everyone |
Launch | DNS change, redirect checks, sitemap submission | Developer and SEO |
Days 1 to 7 | Daily 404, indexing and tracking checks | SEO |
Week 4 | Non-brand review against baseline | SEO |
Week 8 | Recovery review and redirect gap fixes | SEO |
Week 12 | Full report: non-brand clicks, organic revenue, lessons | SEO |
Tool stack
Job | Tools |
|---|---|
Identify the current stack | BuiltWith, Wappalyzer |
Crawl, compare and test redirects | Screaming Frog, Sitebulb |
Search and traffic data | Google Search Console, GA4, Looker Studio |
Backlinks | Ahrefs, Semrush |
Data migration | Matrixify, Cart2Cart, LitExtension |
Filtering | Shopify Search & Discovery, Searchanise, Boost |
Reviews | Judge.me, Yotpo, Okendo |
Markets and translation | Shopify Markets, Translate & Adapt |
Product feeds | Google & YouTube app, Channable, Feedonomics |
Frequently asked questions
Can I migrate from Magento to Shopify in phases?
Phasing by URL path means some paths sit on Magento and others on Shopify, which needs a reverse proxy in front of both. Shopify merchants report certificate and connection problems with proxies in front of a store (Shopify Community). If you must phase, split by market or domain, not by path.
Do I need Shopify Plus to protect my SEO?
No. The SEO mechanics are the same on every plan, and the 2,048 variant limit applies across plans. The difference that matters is redirect capacity: 100,000 on standard plans and 20,000,000 on Plus (Shopify Help Center). Count the rows in your map before choosing.
Can I change domain during the same migration?
Avoid it. A domain change adds redirects from the old domain, which Google says to keep for at least 180 days and ideally longer, plus the Change of Address tool in Search Console, which only works at domain level and lasts 180 days (Google Search Console Help). Doing both at once makes any loss impossible to trace. Move the domain separately.
Can I keep my Magento .html URLs on Shopify?
Not sensibly. Shopify’s /products/, /collections/ and /pages/ prefixes are fixed, and the Help Center says a .html URL cannot be redirected to the same URL without the extension. Redirect each .html path to its explicit Shopify destination and update internal links to the new URLs (Shopify Help Center).
Next steps
If you are planning a Magento to Shopify migration and want a second pair of eyes on the redirect map before launch, I review the url_rewrite export, the map and your Search Console data, and send back a prioritised list of the URLs and patterns most at risk.
Summary
A Magento to Shopify SEO migration is won in the redirect map and the filter page decisions, not on launch day. Build the baseline, pull the url_rewrite table and extension list from Magento itself, and map every URL once, in one hop, to a relevant page. Rebuild the filter pages that earn traffic as collections, keep content and structured data like for like, launch outside your peak, and judge the result on non-brand clicks and organic revenue over 12 weeks.
Kev Wiles - Fractional Shopify SEO Consultant © 2026














