From disconnected work to trusted systems.
For operators who inherited a system nobody can safely change: get an honest read before you rebuild anything. That includes the cases where the right move is to keep what you already have.
Any of these sound familiar?
One person understands the system. Everyone else guesses.
It still runs. Nobody knows for how much longer.
The same data gets typed in two places.
The reports do not match the work.
You cannot safely add a portal, an integration, or mobile access without understanding the risk.
An audit or an acquisition exposed dependencies nobody knew about.
The five paths
Not every legacy system should be replaced. Most paths cost less than a rebuild.
Keep
~$0Change nothing structural. Document it.
Right when
Stable, the expert is not leaving, no new integration is needed.
Wrong when
One person is a single point of failure.
Stabilize
LowCut continuity risk: documentation, backups, access, remove the bus factor.
Right when
The system works. The risk is people, not the platform.
Wrong when
The platform itself is failing.
Wrap
MidKeep the core. Add a modern layer: portal, API, mobile, integration.
Right when
The logic is sound but it cannot talk to other systems.
Wrong when
The core is unmaintainable.
Rebuild
Sprint+Replace it as a modern web app. Keep the business rules.
Right when
The platform is a dead end and the rules are worth saving.
Wrong when
A cheaper path solves the real risk.
Replace or retire
VariesMove to off-the-shelf, or shut it down.
Right when
The workflow is now commodity, or the system no longer earns its keep.
Wrong when
Your rules are genuinely custom.
What a diagnostic produces
A sanitized risk register
The shape of what a diagnostic ends in. This is an example, not a real client system.
All schema changes route through one outside contractor. No documentation exists.
Continuity. If they are unavailable, changes stop.
Customer records live in two systems, reconciled by hand each week.
Data integrity. The two copies drift.
Reporting runs on undocumented calculation fields nobody can verify.
Audit. You cannot prove the numbers.
| Finding | Exposure | Severity | Points to |
|---|---|---|---|
| All schema changes route through one outside contractor. No documentation exists. | Continuity. If they are unavailable, changes stop. | High | Stabilize |
| Customer records live in two systems, reconciled by hand each week. | Data integrity. The two copies drift. | Medium | Wrap |
| Reporting runs on undocumented calculation fields nobody can verify. | Audit. You cannot prove the numbers. | High | Rebuild candidate |
The workflows we see most
How the work is done
When the answer is build
An assessment is vendor-neutral. When it points to a rebuild, here is what execution looks like.
What it costs
The prices are here because you should not have to ask for them.
Fit review
FreeA short call. No file upload. We decide together whether there is a fit, and which of the five paths your situation points to.
Operational System Rescue Diagnostic
$7,500 to $15,000We map the system, the risks, and the honest next step. You get the inventory, the workflow map, the report and integration lineage, the risk register, and a keep, stabilize, wrap, rebuild, or replace recommendation.
Legacy Workflow Replacement Sprint
$25,000 to $65,000A bounded build that replaces one workflow at risk. Fixed scope. You own the code.
Larger work
After the diagnosticWe do not quote a rebuild before we understand the system.