---
title: "Legacy migration without stopping the business | QMates"
source: https://qmates.tech/en/services/engineering-transformation/legacy-code-refactoring/legacy-migration-without-downtime
description: "Rewrite or refactor your legacy system? Buried business rules and missing boundaries push teams to rewrite. Migrating in parts keeps the business running."
updated: 2026-11-21
---

Legacy code migration

# We need to change platform, and the business cannot stop

When a technology ages and the business rules live only in the code, a rewrite from scratch looks like the cleanest way out, because it promises to leave everything behind in one go. Yet forgotten rules and a system that cannot be split make it the riskiest choice. Whatever the rewrite misses comes back as a defect in the new system, and the legacy migration stalls halfway.

## Does this sound familiar?

-   The framework or a core library no longer receives updates, and updating it touches many parts of the system at once.
-   The team has been debating for a while whether to rewrite from scratch or refactor, and the debate never reaches a decision.
-   Finding people willing to work on this stack gets harder every year.
-   A migration has already started, and since then the old and new systems have lived side by side with no date for switching off the old one.
-   Nobody knows for certain which business rules live only in the old code.

## Where does the migration of a legacy system stall?

A stagnant technology, rules nobody wrote down and missing boundaries make a rewrite look like the only way out. That is why the rewrite from scratch appears here among the causes, next to the problems it hopes to solve.

1.  ### A technology ageing without updates

    Frameworks and libraries that are no longer maintained become a security risk and a constraint on hiring. Updates put off for years now touch many parts together, and with each delay starting afresh seems simpler.

    Feeds into

    -   

    [

    Handled by

    Legacy Code Refactoring](https://qmates.tech/en/services/engineering-transformation/legacy-code-refactoring)

2.  ### Business rules that exist only in the code

    Years of special cases, fixed one at a time, live in the old code with no tests or documents to describe them. A rewrite has to rediscover every one, and the ones that slip through resurface as defects after the cutover.

    Feeds into

    -   

    [

    Handled by

    Technical Debt Assessment](https://qmates.tech/en/services/engineering-transformation/technical-debt-assessment)

3.  ### A system that cannot be split

    Migrating in parts requires knowing where one part ends and the next begins. If it is unclear where the domain boundaries run, every piece you extract drags dependencies on the rest along with it, and the migration grinds to a halt.

    Feeds into

    -   

    [

    Handled by

    Domain-Driven Design and EventStorming](https://qmates.tech/en/services/engineering-transformation/domain-driven-design-event-storming)

4.  ### Rewriting from scratch, all at once

    Rewriting the system and then switching to the new one in a single day freezes features for a long time, forces you to maintain two systems and concentrates the risk in one moment. A rewrite makes sense when the technology no longer receives security updates, when no part of the system can be isolated or when the business has changed so much that the system is no longer needed. Even then, replacement happens in parts, using the Strangler Fig pattern.

    [

    Handled by

    Legacy Code Refactoring](https://qmates.tech/en/services/engineering-transformation/legacy-code-refactoring)

- [The service that tackles it](https://qmates.tech/en/services/engineering-transformation/legacy-code-refactoring)
- [Not sure which cause weighs most?](https://qmates.tech/en/services/engineering-transformation/technical-debt-assessment)

## What changes when the migration proceeds in parts

The new system grows alongside the old one and takes over one function at a time, each protected by tests written before the move. The team keeps shipping features while it migrates, and the old code is switched off piece by piece, once nobody uses it any more.

On the roadmap, the migration shows up alongside new features as a series of steps, each naming the function to move. A step can be stopped or corrected without losing the work already done. The rules that lived only in the old code now sit in the tests, and move to the new system with everything else.

- [SureVIVE: modernising the architecture without stopping delivery](https://qmates.tech/en/case-studies/surevive)
