Rescue or rebuild
Rescue or rebuild? How to decide what to do with a struggling software project
Your software is late, fragile or expensive to change, and someone has suggested starting over. Before you decide, look at what a rebuild really costs and what the evidence says about your own code.
Updated
Why a rebuild is rarely the safe choice
A rebuild feels clean: new code, a new team, none of the old problems. But the old system holds years of business rules, edge cases and fixes that nobody wrote down. A rebuild has to find and reproduce all of them, while the old system still needs support and your users wait for new features.
Many struggling projects are not stuck because the code is beyond repair. They are stuck because of a broken release process, tests that nobody trusts, unclear ownership or missing documentation. A rebuild does not fix any of those by itself, and the new code can end up in the same state.
Which signs do you see?
Not sure the project needs help at all? Start with the signs a software project needs a rescue.
Signs that rescuing the code is the better path
- The product works for its users, but every change is slow or risky.
- The problems are in the process: releases, tests, environments or access.
- The framework is old, but it still has a supported upgrade path.
- The business rules live in the code and nowhere else.
- You cannot stop new features for the months a rebuild takes.
Signs that a rebuild may be justified
- The platform or language has no supported upgrade path, and nobody will support it.
- The product must now do something the current architecture cannot do.
- The code base is small enough that writing it again costs less than learning it.
- Security or compliance problems are built into the design, not only into parts of the code.
Decide from evidence, not opinion
Both choices are expensive to get wrong. A vendor that has not read your code can only guess which one is right.
An independent technical assessment reviews the code, the architecture, the process, the release pipeline and the security. It says what to fix, in what order, and what the work will cost. Briebug's Technical Assessment takes about 1 week for a fixed fee of $10K, and the report is yours to take to anyone.
PSIA / SnowPro
Education, certification and membership platform for ski instructors
Upgraded, not rebuilt: Angular 7 to 17, .NET Core 3.1 to .NET 8, working test suites, and repositories, pipelines and documents in an Azure DevOps environment PSIA owns.
Read the PSIA / SnowPro case studyCommon questions
Is it cheaper to rebuild software from scratch?
Rarely, once you count everything. A rebuild has to find and reproduce every business rule in the old system, and the old system needs support until the new one replaces it. An assessment gives you the numbers for your own project before you choose.
How do I know if our code is worth keeping?
Have someone independent read it. An assessment looks at the code, the architecture, the tests, the release pipeline and the security, and says which parts are sound and which are not.
Can we rescue part of the system and rebuild the rest?
Yes. The triage report says which parts to fix, which to replace, and in what order, so you can replace the worst parts while the rest keeps running.
How long does a technical assessment take, and what does it cost?
About 1 week, for a fixed fee of $10K. You get a comprehensive report you can take to anyone, a triage report of what to fix in what order, and a milestone-based estimate.
Find out what your code is worth before you decide.
Book a short call with a senior engineer. We will scope the assessment and confirm the fixed fee before you commit.