Legacy application modernisation

Keep what works. Rebuild what’s underneath.

Keep the workflows people rely on while addressing unsupported software, ageing servers and fragile integrations. We assess the application, agree what must be preserved and plan the move to a supported stack in Azure or AWS.

Since the 1990s Experience across generations of applications and infrastructure.

Talk to an engineer Start with a scoped review
Engineering focus
01Understand the system
02Prove its behaviour
03Rebuild and move it
Possible applicationCharities & membership organisations

Replace an unsupported membership database while preserving agreed record history, renewal rules and links to other systems. Reconcile the data and rehearse cut-over before staff move across. Explore this sector.

Is this you?

The system works. Its platform is falling behind.

An application can remain useful long after its platform becomes difficult to support. These are common reasons to review the options.

Unsupported software.

The application, its database or its server no longer receives security updates.

Ageing hardware.

The application depends on an ageing server that is difficult to maintain or replace.

A vendor that has moved on.

The supplier has gone, the product is discontinued or continued support is no longer practical.

One person who understands it.

Essential knowledge sits with one person or is no longer available.

Locked out of modern tools.

It cannot connect to Microsoft 365, other systems or AI, so people rekey data by hand.

Security gaps.

Unsupported components or outdated controls leave gaps that need assessment and an agreed remediation plan.

The method

Six steps. Evidence at each stage.

Engineer-led and AI-assisted. AI helps with reading, testing and mapping. Engineers review the proposed changes against agreed behaviour and acceptance tests.

  1. Understand

    AI-assisted analysis of the code, database and screens to identify business rules that have not been documented.

    • Code analysis
    • Data mapping
    • Business rules
    What we do
    We read the code, schema and data with AI assistance, then talk to the people who use the system every day.
    What you get
    A documented model of the system and its rules, before anything changes.
  2. Prove

    Tests that capture the agreed workflows and important edge cases, so the new version can be held to them.

    • Characterisation tests
    • Reference data
    • Acceptance criteria
    What we do
    We record real inputs and outputs and turn them into automated tests.
    What you get
    Evidence that the agreed behaviour passes its tests, with any differences recorded.
  3. Rebuild

    A supported stack with the workflows and screens agreed for preservation.

    • .NET
    • SQL Server or Azure SQL
    • Web front end
    What we do
    We rebuild module by module, running the parity tests at every step.
    What you get
    The agreed application behaviour on a supported platform, with the differences and support lifecycle documented.
  4. Move

    Data migrated into Azure or AWS against agreed reconciliation rules.

    • Azure
    • AWS
    • UK regions
    What we do
    We rehearse the migration, reconcile the agreed data and investigate discrepancies. The rollback or recovery plan is agreed before cut-over.
    What you get
    A planned cut-over, with retention of the old system and the recovery route agreed before sign-off.
  5. Run

    Agree security, backup and monitoring before go-live.

    • Monitoring
    • Backups
    • Patching
    What we do
    We run it under an agreed service scope, or hand it over with the code, documentation, tests and runbooks specified for your team.
    What you get
    A supported system with a clear owner.
  6. Enhance

    Add useful AI capabilities as the system evolves.

    • Search
    • Summaries
    • Agents
    What we do
    We add AI where it earns its place, such as search across records or drafted summaries.
    What you get
    Targeted improvements, tested against the workflow and introduced when there is a useful case for them.
Select a step to see what happens and what you receive.
After go-live

Your system. Your choice.

Agree who will operate the modernised system and what a future handover would require.

Our team runs it.

Hosted in Azure or AWS, in UK regions, and run by Toltec under a service agreement: monitored, patched and backed up.

Your team runs it.

Handed over in your own cloud tenancy, with the code, documentation, tests and runbooks your team needs to run it.

Systems we modernise

Old platforms. Familiar territory.

Our experience across generations of systems helps us understand their dependencies and the business rules that have accumulated around them. Explore our sector experience.

Access databases.

Departmental systems that outgrew the desktop.

Bespoke .NET and Windows applications.

Line-of-business software built years ago, still doing its job.

Legacy CRMs.

Customer systems the organisation depends on but the vendor no longer develops.

Unsupported SQL Server and Oracle.

Databases on versions past the end of support.

Discontinued SaaS products.

Assess the available data and workflows, then agree what can be rebuilt as a system you own.

Spreadsheet-driven processes.

Critical workbook processes rebuilt as an application with defined rules, access and records of changes.

The first step

An assessment. Then a decision.

Start with a conversation about the application. We then agree a fixed-scope review of its behaviour, data and risks, with a practical plan you can use to decide what happens next.

What it does.

Its modules, integrations and the business rules the code actually enforces.

What it risks.

Support status, security exposure, single points of knowledge and recoverability.

What it needs.

A target architecture in Azure or AWS, a staged plan and the behaviour tests needed to assess it.

See a sample assessment (PDF)
A closer look

Good questions. Straight answers.

Will it look different to our users?

We agree which workflows and screens to preserve, and record any necessary changes before implementation. Acceptance tests check the agreed behaviour.

What if we have no source code?

We assess whether the system can be rebuilt from observable behaviour and accessible data. Agreed behaviour tests define what the replacement must preserve and where evidence is incomplete.

Where will our data live?

In UK regions of Azure or AWS, in your own cloud tenancy or ours, as you choose.

Who owns what?

You always own your data. Ownership of the code is agreed at the start, and the handover option includes all of it.

Talk to an engineer

Tell us about the system you rely on.

Call 0345 257 4447 or send an outline. The first conversation is with an engineer. If an assessment would help, we agree its scope before starting.