A platform migration rarely fails because a CSV file will not upload. It fails when inaccurate product data reaches a live collection, customer accounts do not behave as expected, or an order integration is missed until the warehouse cannot fulfil it. This Shopify data migration guide is built for retailers that need to protect revenue, operational continuity and customer trust while changing platform.
Start with the commercial outcome, not the export
Before mapping a single field, define what the migration must achieve. For a growing direct-to-consumer brand, that may mean better mobile conversion and cleaner merchandising. For an established retailer, it may mean consolidating online and POS stock, replacing manual order processes, or connecting Shopify to an ERP, 3PL, loyalty platform or subscription provider.
That outcome determines the scope. Migrating every historic record is not always the right commercial decision. Ten years of orders may be valuable for reporting, but they can add cost, complexity and unnecessary data handling. Equally, moving only products and customers may create problems for customer service teams if they cannot access order history after launch.
A good discovery phase separates data into three categories: data required to trade on day one, data needed for customer service or compliance, and data that can remain archived in the previous platform. This is where project certainty begins. The migration scope should be itemised, owned and agreed before delivery starts - zero surprises later.
Audit the source data before building Shopify
Most legacy platforms hold more data than the business realises, and not all of it is useful. Duplicate customer records, inconsistent product types, blank variant attributes and retired discount rules are common. Moving poor-quality data into Shopify simply gives the same problems a more modern home.
Run an audit across products, variants, collections, customers, orders, inventory, content and redirects. Review the volume of each dataset, the source of truth and the condition of the fields. If stock is managed in an ERP, for example, that system should remain authoritative rather than being overwritten by a one-off Shopify import.
Product data deserves particular attention. A retailer with 5,000 SKUs may have product titles, tags, options and images distributed across spreadsheets, a PIM and an old ecommerce platform. Decide which source owns each field. Without that decision, teams can spend weeks debating which description, price or image is correct.
Customer migration also requires restraint. Passwords generally cannot be transferred in a usable form from another platform, so account activation or password reset journeys must be planned and tested. Marketing consent must be handled in line with your recorded permissions, not assumed from a customer’s purchase history.
Build a Shopify data migration map
A data map translates legacy fields into Shopify fields and identifies anything that needs transforming. It is both a technical document and an operational agreement: it makes clear what is moving, what is changing and what will not be carried over.
For products, map handles, titles, descriptions, SKUs, barcodes, prices, compare-at prices, weights, options, images, tags, collections, metafields and inventory locations. For customers, map names, email addresses, phone numbers, default addresses, tags and marketing status. For orders, establish which fields must be retained for service, reporting, tax and financial reconciliation.
Do not treat fields with similar names as automatically equivalent. A legacy “category” may need to become a Shopify collection, a product type, a tag or a metafield depending on how the new store will filter, merchandise and report on products. This is one of the most consequential design decisions in a migration because it affects navigation, search, feeds and internal processes after launch.
A migration map should also cover dependencies. If an app, ERP or warehouse system uses SKU values, changing SKU formats mid-project can break fulfilment. If customer tags trigger a loyalty tier or a trade pricing rule, those tags need controlled logic rather than a broad import.
Clean and prepare data in controlled batches
Data cleansing is not glamorous, but it is cheaper before launch than after it. Remove duplicate records, standardise formats and retire values that no longer serve a purpose. Use a consistent approach to product titles, option names, weights, dimensions and image filenames, particularly where product feeds or shipping rules rely on them.
Work in batches rather than attempting one giant import. A smaller product set exposes mapping errors early, while a customer sample reveals whether addresses, consent fields and account invitations behave correctly. Each batch should have a clear record of its source file, transformation rules, import date and validation result.
For larger catalogues, establish rules for exceptions before they appear. What happens to products with no images? Variants without a SKU? Orders linked to deleted customers? A defined exception process prevents a migration team from making inconsistent judgement calls under launch pressure.
Test the migration in Shopify before committing
A test migration is not a cosmetic check. It is a rehearsal of how the business will operate. Import representative data into a development or staging environment, then ask the people who will use the store to test it: ecommerce, merchandising, customer service, finance, fulfilment and retail operations.
Check more than record counts. A store can contain the right number of products and still be commercially wrong. Test whether products appear in the correct collections, filters return sensible results, images are correctly ordered, prices display as intended and stock is accurate at each location.
Customer service should be able to find a migrated customer, understand their status and access the information they need. Finance should be able to reconcile relevant order totals. Warehouse teams should verify that orders flow to the correct fulfilment process and that the integration does not create duplicate or missing orders.
Use a formal acceptance checklist covering at least these areas:
- product, variant and inventory accuracy
- customer records, consent status and account activation
- historic order visibility and financial reconciliation
- ERP, 3PL, POS, loyalty and subscription integrations
- redirects, search behaviour, collection rules and analytics tracking
Plan the cutover around trading reality
The final migration should be planned backwards from launch. Identify the data freeze window, the final export time, the team responsible for each task and the decision-maker who can approve go-live. Avoid treating launch day as the end of the project. It is the point where the plan meets real customers, real stock and real orders.
The right cutover model depends on your operation. A low-volume retailer may accept a short trading pause while final inventory and orders are moved. A high-volume business may need a phased approach, controlled synchronisation or a carefully managed delta import to limit disruption. There is no universal answer, but there should be one accountable plan.
Protect SEO during the move by mapping valuable legacy URLs to their most relevant Shopify destinations. Redirecting every old page to the homepage is quick, but it produces a poor customer journey and can weaken search performance. Preserve page titles, metadata and content where they remain useful, while taking the opportunity to retire thin or obsolete pages.
Set launch-day monitoring in advance. Watch checkout completion, payment outcomes, order creation, inventory updates, shipping integrations and key conversion paths. Keep the former platform available in read-only form for an agreed period if customer service, finance or operations need to reference historic information.
Treat post-launch validation as part of the migration
The first 72 hours after launch deserve structured attention. Reconcile orders and payments, review failed fulfilment messages, check inventory movements and inspect customer enquiries for patterns. If customers are asking where their account history has gone, or warehouse staff are manually correcting stock, resolve the root cause rather than treating each incident as isolated.
This is also the time to assess whether the new data model supports the commercial plan. Can the merchandising team build collections without developer support? Can marketing create useful customer segments? Can operations identify exceptions quickly? Shopify should reduce friction, not merely reproduce the limitations of the old platform.
For complex migrations, a specialist Shopify partner can bring the discipline that internal teams often struggle to create alongside day-to-day trading. PrePixel approaches migration as a controlled delivery programme: clear scope, mapped dependencies, testable acceptance criteria and visible ownership from discovery to post-launch checks.
A successful migration is not judged by whether the import completed. It is judged by whether customers can buy confidently, teams can operate accurately and the business has a cleaner foundation for its next stage of growth.