Skip to content

Legacy system replacement

Most companies do not stay on a twenty-year-old system because they like it. They stay because every replacement quote came back as a two-year project with a seven-figure number attached, and the last attempt did not survive contact with how the business actually works.

You probably recognise

  • A core system nobody will touch because the person who wrote it retired
  • A replacement project that stalled, restarted, or quietly got shelved
  • Critical logic living in spreadsheets that shadow the real system
  • A vendor quote where most of the cost is configuration and discovery

Why these projects fail

Rarely for technical reasons. They fail because the replacement is specified up front, in a document, by people who cannot know everything the old system quietly does. Eighteen months later the gaps surface all at once, during a cutover, with the business on the line. The scale of the bet is what makes it fragile — not the technology.

How we approach it instead

We replace one function at a time, running alongside the existing system rather than instead of it. Receiving first, or labels, or whichever piece is causing the most pain. Each piece goes live on its own, so a wrong assumption costs a week rather than a year. The old system stays running until the last thing it does has somewhere else to live.

What makes this affordable now

The reason these projects were priced in the hundreds of thousands is that they required large teams working for a long time, and most of that work was mechanical — data models, screens, integrations, migrations. That portion compresses dramatically now. The parts that never compressed — understanding your process, getting the domain right, knowing which shortcut becomes a problem in three years — are what we spend the time on.

Scope and timeline. Usually phased. Each phase is scoped and priced on its own, and the first is small on purpose.

Is this the thing costing you the most?

Describe it in a few sentences. We will tell you honestly whether it is worth building, and roughly what it would take.

Start a conversation