EXISTING SOFTWARE

Rebuild or continue your software: when to choose which?

Your software is in use, but every change takes a long time, the developer has left, or nobody dares to touch it. The question follows by itself: have it rebuilt, or keep building on what is there? This is how I make that choice.

The short answer

Rebuilding is rarely the first answer. It costs a lot and also throws away what does work.

Usually the software itself still does what it should. The risk is in everything around it: who can reach it, how a new version goes live and what has not been updated for a while. You solve that by continuing, not by starting over.

Unsure about your own software? Tell me on WhatsApp what is there. You hear what a code review costs before I start.

Signs that continuing is enough

If you mostly recognise these points, the foundation is probably usable.

  • The software still does what it should, but small changes take a long time.
  • The developer who built it has left, or the agency has stopped or become too expensive.
  • Nobody knows exactly how a new version goes live.
  • Updates have not been done for a while.
  • A first version built with AI tools has grown bigger than anyone intended.

Signs that a rebuild is justified

Sometimes rebuilding is the right choice. These are the cases.

  • The technology is no longer supported.
  • The foundation makes every change unreliable: what you adjust in one place breaks in another.
  • The software now has to do something different from what it was built for.
  • The source code can no longer be recovered.

Even then, rebuilding has to cost demonstrably less than repairing. You only know that once someone has read the code.

The third way: in parts

It is not a choice between all or nothing. Modernising or rebuilding one part while the rest keeps running is usually safer than doing everything at once.

For SmartLoxs, for example, I moved the administrative functions out of the main dashboard into a separate admin application. The rest of the platform simply stayed in use.

Six questions you can answer yourself today

  1. Can you reach the source code yourself?
  2. In whose name are the domain, the hosting and the accounts with external services?
  3. Does anyone know how a new version goes live?
  4. Is the technology still supported?
  5. Does the software still do what it was built for?
  6. When was a backup last restored to see whether it works?

If you do not know the answer to several of these, that is no reason to throw everything away. It is a reason to have someone look first, before you decide.

What a code review tells you

Before I change anything, I read the code, the data and the way everything is deployed.

You then get, in plain language, what is there, where the risks are, what needs attention first and roughly what an extension would take. You know what the review costs before it starts.

That is enough to decide on, even if you do not continue with me afterwards.

If you continue: the order

First, access: the code, the hosting, the database, the domain and the external services the software depends on. If something is missing, we find out who has it.

Then whatever carries risk: backups, outdated parts and a reliable way to publish. New features come only after that.

I keep the first change deliberately small. It proves the route to going live works before anything large goes through it.

When I advise against it

If standard software solves the problem more cheaply, if the source code can no longer be recovered, or if rebuilding demonstrably costs less than repairing, you will hear that after the code review.

If an idea is unnecessarily complicated or expensive, I say so before it becomes code.

Unsure about your own software?

Tell me what is there and where it hurts. You hear what a code review costs before I start.

Send a WhatsAppContact form

A link, a screenshot or a voice message is enough. Usually a reply within one business day.

If an idea is unnecessarily complicated or expensive, I say so before it becomes code.