Stories
Field notes on legacy system migration
Grouped by the question you're actually asking — from "is this a problem" through to "who do I hire."
Is this actually a problem?
- Signs your working system has quietly become a liability
A system that still works isn't the same as a system that's still safe to depend on. These are the signs your legacy system has quietly become a liability.
- Why legacy migrations fail: the failure modes nobody budgets for
Most legacy migrations don't fail on the big risks everyone plans for. They fail on the failure modes nobody budgets for — here's what those actually look like.
- Your legacy system is one retirement away from a crisis
Legacy system key person risk is one of the least-tracked liabilities in engineering. Here's how to tell if your system is a single retirement away from a crisis.
- The real price of migration downtime — and how to put a number on it
Migration downtime costs more than the outage window. Here's how to calculate the real cost of downtime before you decide how much risk a cutover is worth taking.
- When "if it ain't broke" starts costing you money
A working legacy system isn't free to keep running. Here's how to spot the point where the hidden cost of legacy systems exceeds the risk of moving them.
How would this even work?
- Cutover day: the runbook that keeps a migration boring
A good cutover runbook makes migration day boring instead of tense. Here's what actually belongs in one, and why boring is the goal.
- Designing a cutover you can safely roll back
A rollback cutover plan is what makes a migration reversible instead of a one-way door. Here's how to design one that actually works under pressure.
- How to prove a data migration is correct before you flip the switch
"It looks right" isn't proof. Here's how to verify a data migration is actually correct before you flip the switch to the new system.
- Zero-downtime database migration with the dual-write pattern
The dual-write pattern is how a zero-downtime database migration gets verified before cutover. Here's how it works and where it tends to go wrong.
- Big-bang vs. strangler-fig: choosing without betting the business
A strangler fig migration and a big-bang cutover solve the same problem with very different risk profiles. Here's how to choose without betting the business on it.
Who do you hire for this?
- How to hire for a migration without getting burned
Hiring the wrong legacy migration consultant is expensive in ways that don't show up until midway through the engagement. Here's what to check before you sign.
- The migration risk audit, and why it comes first
A migration risk audit is the first deliverable in any Marel Labs engagement, not a sales step. Here's what it covers and why it comes before anything else.
- What a legacy migration specialist actually does
A legacy migration specialist is not a generalist contractor. Here's what the work actually involves, day to day, across an engagement.