You've decided to move an old system. Good. Now you face three doors. Most teams walk through the wrong one — and only find out a year and a budget later.
The three options have names that make them sound interchangeable. They aren't. Each one solves a different problem. Pick the one that doesn't match your situation and you'll either overspend or underfix.
Here's how to tell them apart — and how to choose.
Door 1: Move it. Door 2: Fix it. Door 3: Rebuild it.
Rehosting means picking up the old system and setting it down somewhere new — like the cloud — without changing how it works. It's fastest and cheapest today. But you've moved your problems, not fixed them. Good for getting out of an old data center fast when you plan to clean up later.
Refactoring keeps the system but fixes the worst parts: tidy the wiring, replace the rusty pipes. More work than moving, far less than rebuilding. Good when the system is worth keeping but is slowing you down. Rearchitecting knocks it down and builds new — same purpose, new everything. Most expensive, slowest, but the only path that removes old limits for good.
The mistake is matching the wrong effort to the problem
Teams most often rebuild what they should have just moved — paying for a full rearchitect when a clean rehost would have bought three more years of stable operation at a fraction of the cost.
The reverse also happens: moving a system that was actively blocking real business goals, only to spend the next two years doing the work they should have done the first time. Either mistake costs double.
- If the system works fine and the problem is just where it lives — move it.
- If the system is slow or hard to change but still valuable — fix the worst parts.
- If the system is blocking your core goals and patching would just delay pain — rebuild it.
One question cuts through the noise
Ask this: what is this system stopping us from doing? If the answer is nothing important, move it. If it's a few things, clean it up. If it's your whole future, rebuild it.
Match the effort to the problem size. There is no best door — only the right door for what this system is costing you.
Closing view
The goal is not to pick the most ambitious option. The goal is to solve the actual problem without creating a bigger one.
Match the path to the problem, and you only pay once.
