Key takeaways
- An upgrade, a migration and a rebuild are three different projects, with very different effort and risk.
- Replatforming pays off when the platform permanently holds back business goals, not because it is old.
- The ecommerce platform decision starts with requirements and costs, not with picking a system.
Contents
Many online retailers face the same question. The store software has been running for years and has been extended and customized again and again. Every change takes longer, and every update is a risk. At some point someone says: “We need a new store.”
Ecommerce replatforming is one of the biggest decisions in online retail. It ties up budget, people and attention for months. That is why it pays to look closely beforehand: do we really need a new platform, or would a smaller step be enough?
Three paths, three projects
In everyday conversation, the terms get mixed up. Yet an upgrade, a platform migration and a rebuild differ fundamentally in scope, cost and risk.
Three paths to a modern platform
same platform, new versionSmallest intervention. Security and features stay current.
new platform or major versionData, features and integrations move over. The design can stay.
technology, design and processes redoneBiggest lever, biggest risk. Only worthwhile with clear goals.
Upgrade: The platform stays, and the version is updated. Extensions and customizations have to keep pace. This is the smallest step and is often underestimated. Many problems can be solved with a well-maintained, current system.
Migration: The store moves to a new platform or a new major version. Products, customers, orders, content and integrations are transferred. The design can largely stay the same. A typical case: Shopware 5, whose support ended in July 2024.
Rebuild: Technology, design and often processes are rethought. This is the biggest lever, but also the biggest risk. A rebuild only pays off if the goals are clear and the organization can handle the change.
How to tell that a platform change is necessary
A platform migration makes sense when the existing solution permanently blocks core business goals, and when the problems can no longer be solved economically with smaller adjustments. Typical signs:
- Rising operating costs: Licensing, hosting, maintenance and development get more expensive without the store getting any better.
- Lack of security: There are no more updates, or security vulnerabilities can only be closed with great effort.
- Long turnaround times: Simple requirements, such as a new payment method or a new marketplace, take weeks.
- Difficult integrations: The connection to ERP, PIM or marketing tools is error-prone or expensive.
- Dependence on individuals: Only one developer or one agency really understands the system.
A poor design alone, on the other hand, is rarely a reason for replatforming. It can usually be improved on the existing platform too.
The question is not whether your store is old. The question is whether it is holding your business back.
The ecommerce platform decision in five steps
1. Collect goals and problems
What should get better? Write down concrete problems, for example: “Creating new products takes too long” or “B2B customers can’t see individual prices.” Rate each problem by its economic significance.
2. Honestly assess the existing solution
Can the most important problems be solved with an upgrade, better maintenance or individual extensions? Get an independent opinion. Not every agency that wants to build a new store is neutral.
3. Document requirements
Before systems are compared, a clear requirements profile is needed: product range, customer groups, pricing logic, countries and languages, integrations, warehouse and service processes. Without this profile, you end up comparing marketing promises instead of solutions.
4. Compare total cost
Compare not just the project costs but the costs over several years: licenses, hosting, maintenance, extensions, internal working time. A cheap launch can become expensive to run, and vice versa.
5. Plan for risks
Data migration is one of the most sensitive phases of a platform migration. Check beforehand which data is really needed. And build the SEO migration into the plan from the start: if URLs, content or technical signals are not carried over cleanly, visibility on Google can suffer significantly.
Common mistakes
- Starting with the system: The choice of platform comes first, and the requirements are added afterward.
- Changing everything at once: New platform, new design, new processes and a new ERP at the same time. The risk multiplies.
- Carrying over legacy baggage: Every old special-case solution is rebuilt instead of checking whether it is still needed.
- Underestimating internal work: Testing, data maintenance, coordination and training take a lot of time from your own team.
Conclusion
Replatforming is not an end in itself. It is the right call when the existing solution permanently holds the business back and smaller steps are no longer enough. Starting with goals, requirements and total cost leads to a better decision than starting with the question “Which system?”
Often the best solution is a middle path: first stabilize and modernize, then switch in a targeted way. That keeps the store able to act while the foundation for the next step is being laid.
Frequently asked questions
When is an upgrade enough instead of a new store?
When the platform is still maintained, covers the most important requirements, and the problems lie mainly in design, maintenance or individual features.
How long does a store migration take?
That depends heavily on the product range, integrations and special functions. Build in generous buffers, and don’t announce a firm date externally before testing is complete.
How do I avoid losing Google rankings after a platform change?
Record all important URLs beforehand, set up clean redirects, carry over content and metadata, and monitor visibility closely after launch.