Enterprise replatforming
We move enterprise brands off Salesforce Commerce Cloud onto Shopify Plus. Catalogue, customers and order history carry over. Cartridges, controllers and Einstein logic get rebuilt, because they have no Shopify equivalent and pretending otherwise is how migrations slip. You get the rebuild scoped and priced before anyone signs anything.
Why brands are leaving
The platform did not get worse. The economics did. These are the four reasons enterprise teams give us when they start the conversation, in the order they usually raise them.
% of GMV
Salesforce does not publish Commerce Cloud pricing. The licence is commonly a share of gross merchandise value negotiated with a representative, so a good year raises your platform bill automatically. Shopify Plus publishes its entry price and caps the variable fee.
Quarterly
Work that needs a cartridge, a controller change and a deployment window moves at the pace of your release train. Teams tell us features that took a quarter on Commerce Cloud ship in a sprint on Shopify, because most of what they were building is already in the platform.
Scarce
Commerce Cloud work needs people who know cartridges, ISML, SFRA and the Open Commerce API. That pool is small and day rates reflect it. Shopify developers are easier to hire and cost less per hour, which changes what your roadmap costs for the next five years.
Dated
Business Manager carries its Demandware heritage. Merchandising and marketing teams end up filing tickets for changes they should be making themselves. Moving the work back to the people who own it is the benefit brands mention most after launch.
After the move
The point of a replatform is not a tidier admin. It is the work your team stops doing and the money you stop spending to stand still.
Shopify Plus starts at 2,300 dollars a month on a three year term. Once 0.35 percent of your sales exceeds that, roughly 657,000 dollars a month, the fee becomes 0.35 percent, and the platform fee is capped at 40,000 dollars a month. Those are published numbers you can put in a board deck, not a figure you negotiate blind.
Shopify runs the highest volume checkout in commerce and improves it continuously. You extend it with Checkout Extensibility and the Checkout Branding API rather than owning a custom checkout cartridge and the regression testing that comes with it.
Collections, metaobjects and the Search and Discovery app put day to day merchandising in an interface. The changes that needed a developer and a deployment on Commerce Cloud become an afternoon.
Shopify Markets runs multiple countries, currencies and languages from one store. Many Commerce Cloud realm and site configurations collapse into markets on a single store, which removes a whole tier of configuration and cost.
The numbers
These figures come from Shopify, from Salesforce and from public company reporting, not from us. Where a number was commissioned by one of the vendors we say so, because you should weigh it accordingly. Commerce Cloud is a capable platform and Salesforce continues to invest in it. The question is what it costs you to stay.
36%
From an April 2023 like for like study of the two checkouts run by a large management consultancy. The study was commissioned by Shopify, so treat it as a vendor figure. It remains the only public side by side measurement of the two checkouts we are aware of.
80%
Reported by Rainbow Shops in Shopify’s published case study after leaving Commerce Cloud. Their point was as much about structure as size: they pay monthly for what they use instead of committing to three years of estimated volume up front.
$2,300
Per month on a three year term, with the variable fee capped at 40,000 dollars a month. Salesforce publishes no Commerce Cloud price at all. You cannot compare what one vendor will not print, and that asymmetry is itself worth noting during procurement.
3 names
Salesforce bought Demandware in 2016 for 2.8 billion dollars, renamed it Commerce Cloud, and began branding the suite Agentforce Commerce from Dreamforce 2025. Developer documentation still says B2C Commerce. Same platform, and worth knowing when you research vendors.
Set expectations first
Most migration pages sell you a data move. On Commerce Cloud that is the small half of the job. Your catalogue, customers and orders will import cleanly. The behaviour around them was written as cartridge code and has to be built again on Shopify. Scope that honestly and the project is predictable. Scope it as an import and it slips.
A cartridge packages code and data. Nothing on Shopify is shaped like one. A single custom checkout cartridge typically becomes a Shopify Function, an app and a theme change, so counting cartridges is the first thing we do and the number drives the estimate.
ISML templates usually carry business logic alongside presentation. Liquid separates the two, so the storefront is built from scratch rather than translated. That is more work than a port and a better outcome, because the logic ends up somewhere you can find it.
Commerce Cloud syncs the back office with scheduled jobs and pipelines. Shopify uses webhooks and API calls that fire on the event. Every ERP, OMS and PIM link is not repointed, it is rethought, then tested on its own.
Predictive sort, recommenders and the models behind them stay on Commerce Cloud. Shopify has native recommendations and the Search and Discovery app, and there are stronger third party options. Either way the logic is rebuilt and worth testing against your old baseline.
Slots and content assets map onto Shopify sections and metaobjects, or onto a headless CMS if you go that way. The architecture is straightforward. The volume of content to move is usually what people underestimate.
Shopify’s free Store Migration app covers Square, WooCommerce, Etsy, Wix, Amazon, eBay, Clover and Lightspeed. Commerce Cloud is not on that list. Data comes out through Business Manager exports and the commerce APIs, then goes in through scripts we write and test.
Years of Commerce Cloud development accumulate cartridges written to work around gaps Shopify does not have. Reproducing all of them is the fastest way to spend your budget on nothing. We separate the complexity your business actually needs from the complexity you inherited, and the second list is usually longer than anyone expects.
SiteGenesis to SFRA, SFRA to the composable storefront, and now Storefront Next. Each of those was a paid rebuild inside the platform you were already paying for. Shopify ships improvements continuously without asking you to re implement, so this is the last migration of this kind you do.
Scope
Every item below is in the fixed price. If your store needs something that is not here we tell you before you sign, not when we invoice.
Data
The import half of the job. All of this carries over, with the caveats named rather than buried.
Names, SKUs, descriptions, pricing, inventory, images, search engine titles and descriptions. Shopify allows up to 2,048 variants and three options per product. That is the platform ceiling, not the usable one, because many themes, apps and channels still expect 100, so we audit your stack before importing anything above it.
Categories become collections, automatic or manual. Commerce Cloud hierarchies are typically deeper than Shopify’s, so the shape is agreed with your merchandising team during mapping rather than flattened by whoever runs the import.
Names, email addresses, phone numbers, full address books, tags and segment membership. Order history stays attached to each record, so accounts feel continuous from the customer side.
Line items, quantities, discounts, tax, totals, both addresses, fulfilment and payment status. Historical orders import as records through the Admin API with customer notifications suppressed, so nobody gets emailed about an order from 2019.
Price books become Shopify price lists. Customer groups become segments, or B2B company accounts where wholesale is involved. The entitlement logic behind them is rebuilt rather than exported.
Content assets, landing pages and editorial move with their metadata intact, and every URL gets a destination in the redirect map.
Straightforward codes and price rules become Shopify discounts. Intricate promotion logic and scheduled campaigns get re authored as discounts or Functions, which is a quiet time sink worth scoping.
Realms and sites become Shopify Markets on one store where the separation is regional, or separate stores where the brands are genuinely distinct. We tell you which one your setup is.
Page titles, meta descriptions, canonical intent and structured data carry over so the pages Google already ranks look the same to it after the move.
The honest list
None of this rides along with a catalogue export. Every item is either built again by hand or handled by a deliberate workaround. Scope it explicitly or it becomes a surprise in week thirty.
Pricing engines, merchandising rules, custom checkout steps and integration cartridges. Some become Shopify Functions, some become apps, some become metaobjects, and a few need genuine custom development. The audit tells you which is which.
Every OCAPI and SCAPI connection to an ERP, OMS, PIM or CDP is rebuilt on Shopify’s Admin GraphQL and Storefront APIs. Enterprise stores typically have between eight and fifteen of these and each is its own piece of work.
Predictive sort and the recommenders do not transfer. Shopify has native recommendations, and Search and Discovery covers a lot, but per shopper ordering needs a dedicated tool. We rebuild it and test against your Commerce Cloud baseline rather than assuming parity.
Internal tools your operations team built over years. Shopify admin is more opinionated, so these are rebuilt as Admin UI extensions or small internal apps. Budget team training as a real workstream, not a handover email.
Abandoned cart, browse abandonment and post purchase journeys are wired to Commerce Cloud events. You either rebuild the event layer to keep Marketing Cloud, move to Klaviyo, or run both during transition. All three are valid and the choice is commercial, not technical.
Passwords are encrypted outside the platform and never transfer from anywhere. On a store opened today the imported profile is already an account and customers sign in with the same email and a one time six digit code. Expect a small login dip in week one and a short support note handles it.
Mapping
Every Commerce Cloud store has a stack of custom and vendor cartridges. Here is where the common ones land on Shopify. The mapping is rarely one to one, which is exactly why the audit comes before the quote.
Shopify Functions cover discount, delivery and payment customisation on Plus. Account level pricing moves to B2B price lists. Complex tiered logic is where Functions earn their place.
The Search and Discovery app is free and handles filters, synonyms and search term boosts. Catalogues that need weighted ranking go to Boost, Klevu or Algolia, and that is a real monthly cost to put in the budget.
Shopify Tax and Shopify Protect are native. Avalara and TaxJar are there if your finance team already depends on them, and Signifyd or NoFraud where fraud rules are bespoke.
Shopify Payments is native and removes the third party transaction fee. Adyen, Stripe, Braintree and the rest are supported where a gateway relationship is contractual.
Delivery customisation Functions handle rate and method logic. Carrier and fulfilment integrations reconnect through apps or a custom app against the Admin API.
Manhattan, Fluent and Salesforce Order Management all have a Shopify path. Each rebuild is a discrete project and this is usually the longest pole in an enterprise move.
The capability migrations lose
This is the row on the parity spreadsheet that comes back blank, and it is worth more attention than it usually gets. Commerce Cloud sorting rules weight several attributes at once. Shopify’s native collection sort uses one criterion at a time. Everything looks fine at launch, because the collections exist and hold the right products. Category page conversion just sits lower, and it takes a quarter to work out why.
Shopify sorts a collection by relevance, best selling, title, price, creation date or manual order. Pick best selling and three year old products hold the top slots. Pick newest and you bury what converts. On Commerce Cloud you weighted both and let stock depth break the tie. That is the gap.
Product boosts in Search and Discovery are assigned per search term. Most catalogue browsing happens on collection pages, where those boosts do not reach. Worth knowing before it is presented as parity.
Commerce Cloud teams are used to proving a sort change helped. Shopify has no native A/B test for collection order, so without a testing layer a drop in category page conversion is very hard to attribute to anything.
We ask two questions early. How many business signals drive your sort order today, and who owns merchandising after launch. If the answer is more than one signal, and merchandisers own it, rebuilding the logic in code recreates the developer dependency you are leaving. That is a design decision and it belongs in the scope.
Free resource
The audit we run before quoting an enterprise migration. Cartridge inventory, integration surface, data volumes, redirect scope and the merchandising questions most parity spreadsheets miss. Use it with us or with anyone else.
No payment details. We send the document.
How it runs
Six phases. Enterprise moves run in months, not weeks, and the cartridge count and integration surface set the calendar. Confirm your Salesforce exit notice window during discovery, because it decides the date more often than any technical detail does.
We inventory cartridges, controllers, ISML customisation, every API integration, Business Manager workflows, content slots, Einstein rules and your URL structure. The parity spreadsheet gets filled in properly, including the rows that come back blank.
Commerce Cloud schemas get mapped to Shopify equivalents field by field. Custom attributes are assigned to metafields or metaobjects now, not improvised later. Extract scripts are written and run against a sample.
The storefront is built on Shopify rather than ported. Checkout behaviour goes into Functions and Checkout Extensibility. Every back office connection is rebuilt against the event model.
We crawl the live Commerce Cloud site and build a redirect map covering every URL that returns 200. Titles, descriptions and structured data carry across. Internal links are repointed before launch, not after.
Full data migration to a staging store, then testing against the parity document rather than against a vague sense of done. Orders, checkout, integrations, markets, currencies and B2B flows all get exercised.
A short content freeze, a delta import to catch orders and customers created in the gap, then DNS. Where B2B is involved we run a controlled rollout to selected accounts first, because real buyers find pricing and permission problems that testing does not, while the blast radius is still small. Redirects go live with the switch.
Ready to scope it?
Get a migration scopeIn practice
Scenario types rather than named clients, because our enterprise work sits under agreements we do not break for a marketing page. On a call we will talk you through comparable projects properly.
Premium apparel, multi market
Four Commerce Cloud sites serving four regions, each with its own catalogue configuration and its own maintenance overhead. Consolidated onto a single Shopify store with Markets handling currency, language and regional pricing. Around forty cartridges audited, eleven rebuilt as Functions or apps, the rest retired because nothing depended on them.
One store replacing four site configurations
Beauty, high SKU count
A deep catalogue with heavy variant structures and nine years of order history the finance team needed kept intact for lifetime value reporting. Extract through Business Manager and the commerce APIs, import through bulk operations, reconciled record by record against source totals before sign off.
Nine years of order history preserved
Wholesale and direct to consumer
Commerce Cloud customer groups with tiered and contract pricing rebuilt as Shopify B2B company accounts and price lists. The catalogue moved cleanly, the entitlement logic was re architected rather than copied, and the wholesale team kept their ordering workflow.
Wholesale and retail on one platform
Organic traffic
Commerce Cloud lets you shape storefront URLs. Shopify uses fixed paths, slash products and slash collections. Almost none of your old URLs survive unchanged, which means every one of them needs an explicit redirect. This is the cheapest part of a migration to get right and the most expensive to skip.
We crawl your live site and export every URL returning 200, then map each to its Shopify equivalent. Shopify allows 100,000 redirects on standard plans and 20 million on Plus, imported in bulk, so volume is not the constraint. Completeness is.
Shopify redirects only trigger on a 404, and they are explicit mappings rather than wildcard rules. Some paths cannot be redirected from at all. We build around both rules and the map goes live at the same moment DNS switches.
Titles, meta descriptions and product, breadcrumb and FAQ structured data move with the new URLs. We do not rewrite copy Google already likes just because the platform changed.
Rankings fluctuate while Google recrawls. That is normal for any replatform and a complete redirect map is what holds equity through the window. Migrations that lose traffic permanently almost always skipped or botched this step.
Which Shopify
Two decisions get made during discovery. Which Shopify you land on, and whether the storefront is Liquid or headless. Coming off a composable storefront does not oblige you to stay headless.
Where most Commerce Cloud brands land. Checkout Extensibility across the information, shipping and payment pages, the Checkout Branding API, Shopify Functions, unlimited B2B catalogues and price lists assigned to companies, and 20 million redirects. Published entry price, capped variable fee.
The default unless there is a specific reason otherwise. Fastest to launch, lowest to build and lowest to run, and it covers the large majority of storefronts. Teams arriving from PWA Kit often assume this is a step down. Usually it is not.
Shopify’s own headless framework with hosting included. It fits teams coming from a composable storefront who want full front end control and have the people to maintain it. Higher to build and higher to run, so it should be a decision rather than a habit.
Budget
We do not publish a price for enterprise replatforming, because a number quoted before the cartridge audit is a number that gets revised. These five things set it, and you will know your figure before you commit.
The single biggest driver. Enterprise stores commonly run dozens of custom and vendor cartridges. Some retire, some become configuration, some need real development. Counting them is week one work and it moves the estimate more than anything else.
Each ERP, OMS, PIM, CDP and middleware link is its own rebuild with its own testing. Eight integrations and fifteen integrations are different projects even when the catalogue is identical.
SKU count matters less than structure. Deep hierarchies, heavy variant matrices, many price books and years of order history all add mapping and reconciliation time.
A native theme and a headless build are materially different budgets, to build and to run. We make the recommendation on your requirements, not on what is more billable.
Marketing Cloud journeys, Data Cloud profiles and Salesforce Order Management each need a decision and usually integration work. Your renewal date matters just as much: Commerce Cloud agreements run on multi year terms with volume commitments, so we sequence the migration against the exit notice window rather than paying for two platforms at once.
Also on Shopify
Migration is usually the start. These are the other places we work.
Adobe Commerce and Magento Open Source migrations, including extension replacement and full order history.
Catalogue, customers and orders moved off BigCommerce, with app replacements mapped before the build.
WordPress and WooCommerce stores moved onto Shopify with the blog and its rankings intact.
Custom theme work, Shopify Functions, custom apps and Checkout Extensibility for Plus stores.
Research, design and testing on stores that already sell, aimed at the numbers rather than the aesthetic.
Marketplaces, retail and social channels connected to one Shopify catalogue and one inventory source.
Why Speed Plus
Enterprise replatforming goes wrong in predictable ways. These four are how we stop it, and they are in the contract rather than on a slide.
Data handling
You are handing over the catalogue, the customer list and years of order history for an enterprise store. Here is exactly how that is handled.
We ask for the narrowest Business Manager and Shopify permissions that let the work happen, named to individuals, and we hand them back the day the project closes.
Extracts move over encrypted connections and sit encrypted on storage restricted to the people on your project. Nothing goes on a laptop or a personal drive.
Access is revoked and credentials destroyed when the project closes. We do not keep a copy of your store for reference and we do not reuse your data for anything.
Commerce Cloud stays live and untouched through cutover. If something is wrong at the switch we point DNS back and you carry on selling while we fix it.
Signed before you send us anything. Catalogue, pricing, customer data and commercial terms stay between us, including the fact that you are considering a move at all.
Questions
The questions enterprise teams actually ask, answered without the sales gloss. If yours is not here, ask it on the form below and you get a real answer rather than a callback.
Yes, all three are the same platform. Salesforce acquired Demandware in 2016 for 2.8 billion dollars and renamed it Commerce Cloud. From Dreamforce 2025 Salesforce began branding the suite Agentforce Commerce, while developer documentation still refers to B2C Commerce. Your contract may say any of them. The migration playbook is identical. Salesforce B2B Commerce, formerly CloudCraze, is a separate product with its own data model.
No. Shopify’s free Store Migration app supports Square, WooCommerce, Etsy, Wix, Amazon, eBay, Clover and Lightspeed. Commerce Cloud is not on that list and there is no dominant off the shelf tool the way there is for Magento. Data comes out through Business Manager exports and the commerce APIs, and goes in through migration scripts written for your store and tested on a sample before the full run.
Enterprise moves run in months rather than weeks, and the cartridge count and integration surface set the calendar more than SKU count does. A lean store with few custom cartridges and clean data is materially faster than one with dozens of cartridges and fifteen API integrations. We give you a date after the audit, not before it, and we confirm your Salesforce exit notice window during discovery because that often decides the schedule.
We quote a fixed price after the cartridge and integration audit, because a number given before that audit is a guess that gets revised upward. Five things set it: how many cartridges you run, how many integrations need rebuilding, how your catalogue is structured, whether you go Liquid or headless, and how deep your Marketing Cloud and Data Cloud dependencies run. You see the figure before you commit.
Not with a complete redirect map. Commerce Cloud lets you shape URLs and Shopify uses fixed paths, so almost none of your old URLs survive unchanged and every one needs an explicit redirect. We crawl your live site, map each URL returning 200 to its Shopify destination, and put the map live at the same moment DNS switches. Expect fluctuation while Google recrawls, which is normal. Permanent losses come from gaps in the map, not from the move itself.
They are rebuilt, because nothing on Shopify is shaped like a cartridge. The logic splits by type: discount, shipping and validation rules become Shopify Functions, storefront behaviour becomes theme code or an app, and scheduled jobs become webhook driven services. One cartridge rarely becomes one Shopify object, which is why the audit comes first and why the count drives the estimate.
Einstein does not transfer. Shopify has a native product recommendations API that covers standard cases, and the free Search and Discovery app handles filters and search term boosts. For per shopper ordering and more sophisticated personalisation you need a dedicated app, which is a real monthly cost. We rebuild the logic during the project and test it against your Commerce Cloud baseline rather than assuming it matches.
Not natively, and this is the honest answer most migration pages skip. Commerce Cloud sorting rules weight several attributes together. Shopify sorts a collection by one criterion at a time: relevance, best selling, title, price, creation date or manual order. Search and Discovery boosts help, but they are assigned per search term and apply to search results rather than collection pages. If more than one business signal drives your ordering today, you either accept a simpler model, build the logic in code, or add a merchandising layer. We settle that in planning because it is a design decision, not a configuration task.
They become one of two shapes. Where the separation is regional, Shopify Markets runs multiple countries, currencies and languages from a single store and replaces many realm configurations outright. Where the brands and catalogues are genuinely distinct, separate stores is the honest answer and it carries real cost and operational consequences. We tell you which one your setup is during the audit rather than defaulting to whichever is easier to build.
Customer groups become Shopify B2B company accounts and price books become price lists. B2B is available from the Basic plan upward, but the heavy pieces, unlimited catalogues and price lists assigned directly to companies, are Plus only. Most Commerce Cloud brands are on Plus anyway. The catalogue moves cleanly. The entitlement logic behind the groups is re architected, not copied, so plan it as a build.
Journeys wired to Commerce Cloud events, abandoned cart, browse abandonment and post purchase, need a decision. You can rebuild the event layer on Shopify data to keep Marketing Cloud, move to Klaviyo, or run both in parallel during transition. All three work. Which is right depends on how deep the journeys run and whether your marketing team wants to change tools at the same time as the platform, which is often one change too many.
Do not assume it keeps working untouched. If Data Cloud holds your unified customer profiles you either rebuild the Shopify event integration into it, replace it with a customer data platform piped to Shopify, or move the segmentation function to your marketing tool. Each is integration work and each belongs in the scope with a number against it.
No, and they never transfer from any platform, because passwords are encrypted outside it. Only the customer records import. On a Shopify store opened today the imported profile is already an account and the customer signs in with the same email address and a one time six digit code rather than setting a password. Order history stays attached, so accounts feel continuous. Expect a small login dip in week one and send a short note explaining the new sign in.
Yes. Historical orders import as records through the Admin API with customer notifications suppressed, so nobody is emailed about an order from years ago. One thing that trips teams up: the Orders API returns only the last 60 days by default. That is a read default you filter around, not a cap on what you can import. Plan how refunds and fulfilment work for pre migration orders, because imported records are not live transactions.
Shopify allows up to 2,048 variants and three options per product. That is the platform ceiling rather than the usable one, because plenty of themes, app extensions and sales channels still expect 100, and theme loops are capped lower. We audit your storefront stack before importing anything above that, and products with very wide matrices sometimes get restructured instead.
Coming from PWA Kit or a composable storefront does not oblige you to stay headless. A native Liquid theme is faster to launch, cheaper to build and cheaper to run, and it covers the large majority of storefronts. Hydrogen fits teams who want full front end control and have the people to maintain it. We make the recommendation on your requirements and on who maintains the storefront afterwards, not on what is more billable.
Yes. Commerce Cloud stays live and taking orders until the new store is signed off. We build and test in parallel, then run a short content freeze, take a delta import to catch orders and customers created in the gap, and switch DNS. Planned downtime is zero minutes and Commerce Cloud stays untouched so a rollback is real rather than theoretical.
When there is no forcing function. If you are mid contract with a long exit notice window, inside a peak sales period, or your store leans on Commerce Cloud behaviour with no clean Shopify path yet, a replatform will cost more than it returns this quarter. A contract renewal, a cost trigger or a capability you are blocked on is what makes the timing right. We will tell you to wait if waiting is the right answer.
Hypercare while the integration surface settles, which on a Commerce Cloud move is larger than on most migrations. We monitor orders, checkout and every rebuilt integration, watch Search Console coverage daily for crawl errors and soft 404s, and fix anything important immediately rather than batching it. Then we hand over documentation, scripts and code that belong to you.
We tell you during planning, in writing, while you can still act on it. Some Commerce Cloud behaviour has no good Shopify answer and some of it was never worth what it cost you to run. Separating those two is most of the value of the audit. A migration partner who says everything maps one to one has not looked at your cartridges yet.
You do. Theme code, custom apps, migration scripts and documentation are handed over and they are yours. There is no proprietary framework you have to keep paying us to touch, which rather defeats the point of leaving a platform that held your logic hostage.
Insights
Posts are on the way.
Start here
Tell us what you are running and we will come back with a cartridge and integration audit, a scope and a fixed price. No obligation, and no call required before you see a number.
Every Magento version, including Magento 1 and Adobe Commerce
Send us your cartridge list and your integration inventory. You get a scope, a timeline and a fixed price back, and an honest answer about whether now is the right moment.
