Guides

Can a legacy system without source code be saved?

Yes, but first separate three different jobs: takeover, migration and rebuild. The cost, risk and timetable are very different.

Definition

Legacy system takeover means the existing system keeps running while a new team takes over accounts, data, day-to-day maintenance and limited changes. It is about taking over, not rewriting.

Four paths with different risk and cost

Not every old system needs to be thrown away. Decide which path pays off first.

  • Repair: fix only what is broken — quickest, but limited by the original design
  • Take over: run and maintain it, with small changes, no rewrite — fits systems that still cope
  • Migrate: change platform but keep the functions; export and restore are the critical part
  • Phased rebuild: keep the core and replace it step by step — lowest downtime risk

Six things the assessment must cover

A quote before looking at these six is a guess.

  • Accounts and permissions: who holds administrator access
  • Whether source code, licences and vendor support exist
  • Database type, size and version
  • Export capability: can everything be extracted and restored
  • APIs or integration points with other systems
  • Data-protection and permission requirements (who may see customer data)

The honest boundary when there is no source code

Without source code we can take over operations, migrate the data and rebuild. What we cannot do is promise to modify the original system.

So treat “we can definitely change it, a few days” from anyone who has not seen your system as a question mark. The right order is a health check first, then a written statement of what is and is not possible.

When rebuilding beats repairing

In these situations, repairing usually gets more expensive every year:

  • Every change needs a workaround and extra manual steps
  • The data does not reconcile and people check it by hand
  • The vendor no longer supports it and security updates have stopped
  • The business process has moved on and the system cannot follow

What a takeover actually involves

Four steps, each with something in writing.

  • Survey and health check (accounts, source, database, export, APIs, data permissions)
  • Written recommendation: repair, take over, migrate or phased rebuild — with risks and costs
  • Scope and acceptance: must-keep features, acceptance criteria, stages
  • Staged delivery and handover notes, with monthly maintenance available afterwards

Frequently asked questions

Can the data be moved out without source code?

Usually yes, depending on the database and export options. The approach is to test export and restore first, and only migrate once the data is proven complete.

Does a takeover require downtime?

Normally not a single big cutover. It can be staged: take over operations first, then replace parts one at a time, keeping the risk contained.

How long does a rebuild take?

It depends on scope and the features that must be kept, so an assessment and written scope come first. Nobody should quote a timeline before seeing the system.

Is your old system worth saving?

Book a legacy systems health check — bring what you know: who built it, source code or not, which database.