Migrating to Shopware without losing revenue
Migrations rarely fail on technology. The products come across, the customers come across, the shop goes live and looks better than ever. Six weeks later organic traffic is down thirty percent and nobody can point to exactly why. That is the real risk of replatforming: not that it does not work, but that you lose visibility and convenience on the way.
Anyone moving from Magento to Shopware, or from Shopware 5 to 6, Shopify to Shopware or Lightspeed to Shopware, is moving far more than data. You are moving URLs that Google has been used to for years, customer accounts with passwords and pricing agreements, and working practices in your sales office that were never written down anywhere. Each of those three is a place where revenue can leak away.
.png)
Why companies switch
There is almost never a single trigger. Usually a few reasons stack up until standing still costs more than moving.
- End of life: no more security updates, so a deadline rather than a wish list
- Accumulated custom code: nobody dares to update anything
- Licence and maintenance costs: the bill grows faster than your turnover
- Slow development: marketing waits on developers
- B2B requirements that do not fit: custom work for what should be standard
That last one comes up most often at wholesalers and manufacturers. Customer-specific pricing, quote flows and multiple users per account are custom work on your current platform, while Shopware treats them as standard. What that covers is set out on the page about the Shopware B2B webshop.
Shopware is also a German platform and the market leader in Germany, which makes it attractive for Dutch companies exporting into the DACH region. It runs on-premise, so you keep control over hosting, performance and the data it holds.
What really migrates, and where it goes wrong
A migration plan starts with an honest inventory of what moves and in what condition. These are the seven categories to work through, each with the place where it goes wrong in practice.
Products and variants
Variants, media, specifications and translations come across. Clean up dirty product data before the move, not afterwards.
Categories
The tree structure moves without much trouble, but the category copy at the foot of the page is often exactly what that page ranks on.
Customers and accounts
Addresses and customer groups come along, password hashes only where the format allows it. If they cannot, plan a reset email.
Orders and history
Usually moved across read-only, so customers still see their history. How far back you go is a deliberate choice.
Prices and contracts
Volume tiers and contract prices often sit half in the old platform and half in the ERP. Decide first which one is the source.
Content
Blogs, banner positions and landing pages are systematically underestimated, while they carry a good share of your organic traffic.
URLs and redirects
The category most often left until last, and precisely where revenue leaks away after go-live.
Three migration routes
Not every situation calls for the same approach. Memo works with three routes, and choosing between them is the single most important decision in the whole project.
1. A clean rebuild. You rebuild the shop in Shopware 6 and migrate only data, not the old structure. This is the route when custom code has silted up, your data model no longer reflects reality, or your process needs rethinking anyway. It asks more of your organisation, but you come out with an installation you can keep developing for years.
2. A phased migration per sales channel. Instead of moving everything at once, you move per country, brand or audience. One sales channel goes live on Shopware, you measure what happens to traffic and conversion, and only then does the rest follow. This route suits organisations with several storefronts or a high big-bang risk. Shopware supports multiple sales channels on one installation, so the phasing already builds the end state.
3. No migration, but performance work. Sometimes the platform is not the problem. For Kempers.de, a German car recycling business, Memo and Hypernode worked on the performance of the existing webshop rather than a full migration. See the case on that performance work. If your complaints are mainly about speed, stability and peak load, a migration is an expensive detour. The page on Shopware hosting and performance sets out what you can achieve without replatforming.
.png)
The SEO plan decides whether your revenue moves with you
This is where most revenue disappears, and where the least is said during the build. Treat SEO as its own workstream with its own deliverable, not as a checklist for the final week.
The core is the redirect matrix: a line-by-line map of every old URL to its new destination, with a 301. You build that matrix from a crawl of the current site, supplemented with the URLs that carry traffic or impressions in Search Console and your log files. Pages with no logical equivalent get a deliberate decision rather than a redirect to the homepage, because Google treats that as a soft 404.
The plan also covers:
- URL structure: decided before you start building
- Structured data: keeping products, breadcrumbs and reviews
- Measurability: a baseline per template before go-live
- Crawl after go-live: status codes, canonicals, robots.txt and sitemaps
Release 6.7.13.0 brought improvements around robots.txt, canonicals and the Merchant Center connection, which makes the default setup easier. It does not excuse you from the redirect matrix.
A realistic step plan
The order matters. What makes a project heavy is almost never the building itself, but the cleaning up and the deciding that comes before it.
- Inventory and route selection: is migrating actually the answer?
- Data audit and clean-up: products, customers, pricing agreements and content
- Setup and build: sales channels, customer groups, templates, integration layer
- Trial migration: expect more than one round
- Testing with real people: your sales office and a few regular customers
- Redirects and SEO delivery: verified before go-live, not after
- Go-live and aftercare: crawl, monitoring and a comparison against the baseline
We always quote the baseline measurement and the SEO workstream as a separate item. Not because they are optional, but because that way you can see what is being spent on them. Buried inside a project total, it is the first thing to quietly disappear when the schedule gets tight, and that is exactly where revenue is lost in a migration.
Memo has been a Shopware Bronze Partner since 2018, is a Shopware Extension Partner and holds seven valid Shopware 6 certifications in-house. Shop migrations are listed by Shopware as one of Memo's specialisations. What a project costs and what makes it vary sits on the costs page; the overview of all Shopware services is on the hub.
Torn between rebuilding, phasing and fixing performance first? Send us the URL of your current shop and an outline of your integrations. In half an hour we will tell you which of the three routes fits your situation and what your SEO plan needs to contain as a minimum.
Frequently asked questions
Will I lose my Google rankings moving from Magento to Shopware?
Not if the redirect matrix is finished before go-live and you choose your URL structure deliberately. Losses come from missing 301s, redirects to the homepage and pages that quietly disappear. Do expect a few weeks of fluctuation while Google reindexes the new structure.
Can I bring my order history to Shopware 6?
Yes. Orders are usually moved across read-only, so customers can see their history and place repeat orders. How far back you go is a choice: taking everything makes the migration heavier, while in practice customers rarely look back more than a couple of years.
Will customers have to set a new password?
That depends on the format your current platform stores passwords in. Sometimes the hashes can come across, sometimes not. If they cannot, plan a reset email around go-live and explain why in the email, otherwise it reads like phishing and conversion suffers.
Is a phased migration more expensive than moving everything at once?
Usually slightly, because you run two environments alongside each other for a while. In return you spread the risk and can adjust after the first sales channel. For organisations with several brands or countries, that almost always outweighs the extra cost.
What if my real problem is speed rather than the platform?
Then migrating is an expensive detour. At Kempers.de, Memo and Hypernode improved the existing webshop without a full migration. Measure first where the delay comes from: hosting, caching, front end, or an integration that waits on an external system on every page view.