Our previous vendor disappeared and the code cannot be continued
Short answer
What you need first is not a rewrite but certainty: what is actually finished, what is not, and what is worth saving. A two-week code review produces an honest assessment and two options with cost and time estimates — continue or rebuild in stages. You decide after seeing the numbers.
Typical duration: Two-week review · execution depends on the option chosen
Review USD 4,000 · two costed options at the end
What this looks like
- The system runs but nobody dares touch it
- No documentation, and whoever wrote it cannot be reached
- Small changes keep breaking something else
- You are not sure you hold all the code and access that should be yours
- Replacement vendors refuse to continue and recommend a full rebuild
Why this happens
Recommending a full rebuild is the safest answer for a replacement vendor: the job is bigger, the risk to them is smaller, and they never have to understand someone else's decisions. For you it means paying twice for the same thing.
Abandoned code is rarely bad all the way through. Usually only parts are broken, and what is truly missing is the knowledge of why something was built that way — which can be recovered by reading it, not by discarding it.
How we solve it
- 01
Secure what exists
Confirm you actually hold all the code, database, domains, and service access. This is often where missing handovers surface.
- 02
Read and run
The code runs in an isolated environment to see what actually works, not what is claimed to work.
- 03
Honest assessment
Which parts are healthy, which are fragile, which are dangerous. Including a security check, because abandoned code often hides access keys.
- 04
Two costed options
Continue or rebuild in stages, each with cost and time estimates. We say which we recommend and why.
- 05
Execution
Once you choose, the work is done with tests and documentation so this situation cannot repeat.
Numbers from our own work
Code and documentation handed over in full — we lock nothing
The review includes security, because stranded access keys are a common finding
We quote both options, including the one that is worth less to us
Mistakes we keep seeing
- Accepting a full-rebuild recommendation when nobody has read the code
- Starting new work before confirming every access is in your hands
- Leaving the system running without backups because everyone is afraid to touch it
- Hiring one person to maintain a system nobody understands
Questions we are asked most
What if our code really is bad?
We will say so, with reasons. But that verdict comes after reading it — not before.
Does the system have to be taken down during the review?
No. The review runs on a separate copy; your system keeps serving as usual.
We have no documentation at all. Is that a problem?
Common, and not a blocker. Documentation is one of the things the review produces.
Can you just maintain it, without a big project?
Yes. A monthly retainer is available for systems already running.
How do we make sure this does not happen again?
The code and documentation become yours, tests run on every release, and no part depends on a single person.
Updated 29 July 2026 · Neuraltan
If this is happening to you
Tell us the situation. We will say plainly whether this is worth doing now, and how long it takes.