Proven onAP Legislature — paper to digital, no downtime
Risk positionReversible at every step
What it means for your business
Keep the business running
Old and new run side by side until the switch is safe.
Cut maintenance cost
Retire expensive, fragile systems piece by piece.
Ready for AI
Clean data and modern architecture that AI can build on.
Ideal for
Organisations held back by ageing software or spreadsheets.
For your technical team
A routing facade in front of the old system, capabilities extracted one at a time behind it, both running together until the new one has earned the traffic. The legacy system goes dark last, not first.
The problem
Why big-bang migrations fail
The rewrite-and-switch approach asks an organisation to accept all of its risk on one weekend, after a period in which the old system could not be improved. It fails often enough that most institutional buyers have a scar from it.
The roadmap freezes for the duration, so the business falls behind while it waits.
The new system is validated against a specification, not against what the old one really did.
Undocumented behaviour — the exceptions someone hard-coded in 2013 — is discovered on go-live.
Rollback means restoring a backup, so the only path is forward through the failure.
How we do it instead
Every capability is reversible at the facade: if the new path misbehaves, traffic returns to the legacy system in a configuration change rather than a restore.
What you get
Deliverables, not activities.
Each of these is something that exists at the end and that you own.
01
Behaviour archaeology
Establishing what the old system actually does, including the undocumented exceptions, before anything is replaced.
02
The routing facade
A layer in front of both systems that decides per capability — and where needed per user — which one answers.
03
Incremental extraction
Capabilities moved one at a time, lowest risk first, each one independently reversible.
04
Dual-run comparison
Both systems answering the same requests with results compared, so the new path earns traffic on evidence.
05
Data migration with history
Records moved so the past survives the change — history is usually the thing the business cannot lose.
06
Decommissioning
The old system switched off deliberately at the end, once nothing depends on it, with the archive retained.
Technology
What this runs on.
Grouped the way a technical evaluator reads it. Everything listed is in production in work on this site.
01
Routing
API facadePer-capability routingFeature flagsTraffic shadowingInstant rollback
02
Extraction
Service decompositionAnti-corruption layersEvent sourcing where it fitsContract testing
The three things that decide whether this goes well.
Access to the legacy system and, if at all possible, to someone who remembers why it works the way it does.
Agreement on which capability moves first — we will recommend the lowest-risk one.
Tolerance for a period where both systems run. It costs a little more and removes most of the risk.
Questions we get asked
Is this slower than a rewrite?
To the final state, sometimes. To the first improvement your users feel, considerably faster — and the risk profile is not comparable.
What if the legacy system has no API?
Most do not. The facade sits wherever the seam is — database, file drop, screen, message queue — and an anti-corruption layer keeps the old model from leaking into the new one.
Can we stop halfway?
Yes, and some organisations sensibly do. Each extracted capability stands on its own, so a paused programme leaves you better off rather than stranded.
Start here
Let’s build what’s next.
Tell us where your business wants to grow. In one conversation we will show you where AI and software will pay back first, and whether we are the right team to build it.
We look at your goals, your current systems and your biggest time and cost drains, then recommend the first AI or software move with the fastest payback.
AP LegislatureRead the story
Le'Securise OpsRead the story