OmniSense – Understand your legacy system before you change it.

New product by Pretius

OmniSense passively monitors real user sessions on your web application, building a rich knowledge base of how your system actually works — no interviews, no outdated docs, no guesswork. Delivered and supported by Pretius UK, led by Oracle ACE Director Matt Mulvaney.

Book a discovery call

Learn about Pretius UK

Delivered and supported by Pretius UK

OmniSense is a product of Pretius — deployed and supported in the UK by our London-based team led by Oracle ACE Director Matt Mulvaney. Pretius Ltd is a fully registered UK entity (Company No: 147557660), governed by English Law, ensuring local contracts, IR35-compliant engagement, and a data governance posture aligned with UK GDPR from day one.

All captured session data remains within your own infrastructure. Pretius can operate OmniSense without ever accessing your sensitive business data.

Matt Mulvaney

Pretius Ltd CEO / London-based leadership

♠ Oracle ACE Director — 1 of 17 worldwide / 300+ in-house specialists

The Challenge

Legacy systems carry invisible technical debt

Systems built 10–15 years ago continue to run business-critical operations — but the people who built them are long gone, the documentation is incomplete or nonexistent, and no one knows which modules are actually being used.

According to a 2024 McKinsey report, as much as 70% of software used by Fortune 500 companies was developed 20 or more years ago. Gartner projects that by 2025, 40% of IT budgets will go toward maintaining technical debt alone. And the legacy modernisation market — valued at $6.5B in 2024 — is expected to reach $14.7B by 2033, reflecting how urgent this problem has become across industries.

For UK organisations, the stakes are higher still. Operating legacy systems on unsupported platforms is not merely a technical debt issue — under DORA (for financial services firms), FCA operational resilience requirements, and UK GDPR, it can constitute an active compliance liability. Yet the standard alternative — commissioning a full modernisation project without understanding what the system actually does — carries its own enormous risk: scope creep, cost overruns, and loss of business-critical logic that nobody documented.

The compounding pressures UK organisations face every day:

Missing or outdated system documentation.
No test coverage for critical business flows.
Unused modules kept in production, inflating maintenance cost and attack surface.
Impossible to accurately estimate rewrite scope and cost without disrupting operations.
Vendor lock-in with no clear exit path (Oracle Forms, legacy .NET, early Java stacks).
No AI layer — the system is falling behind while competitors modernise.
Original developers are gone, taking institutional knowledge with them.
Regulatory pressure to demonstrate ICT resilience but no documented evidence of what the system does.

Key figures:

15+ – Average age of enterprise legacy applications.
70% – Share of Fortune 500 software built over 20 years ago (McKinsey, 2024).
40–50% – Acceleration in modernisation timelines when AI is applied (McKinsey).
£0 – Upfront cost to install OmniSense and start capturing data.
1–2 weeks – Typical time to first documentation output.

What is OmniSense?

Passive intelligence. Zero disruption.

OmniSense installs a lightweight JavaScript snippet alongside your application. It silently records user sessions — screens, actions, DOM state — and transforms them into structured metadata about your system. Sensitive data is automatically anonymised.

How it works — three pillars:

01. Passive session monitoring

Records real user interactions — no simulation, no assumptions. OmniSense sees what users actually do, not what documentation says they should do.

02. Metadata-first architecture

Raw events are transformed into structured metadata: screens, flows, components, process paths, and branch points. This becomes your system's single source of truth.

03. Data stays on your infrastructure

All captured data remains within your environment. Pretius can operate the tool without ever having access to your sensitive business data — a critical requirement for UK organisations subject to UK GDPR, FCA data handling rules, and sector-specific data sovereignty obligations.

Three outcomes you can act on

Benefit 1

Add AI to your existing system

Deploy AI-powered backoffice tools and an in-system AI assistant without rewriting a single line of code. Test ROI before committing to large-scale investment. For UK enterprises facing pressure from the board to demonstrate AI adoption, OmniSense provides a low-risk, evidence-based path to genuine AI integration — grounded in how your system actually works, not a generic overlay.

Benefit 2

Know the true cost of migration

Instantly scope a rewrite project based on real system behaviour — not assumptions. Generate accurate time, cost, and effort estimates before approaching vendors. For UK IT leaders responsible for business cases, procurement approvals, and governance sign-offs, this evidence-based scoping replaces weeks of expensive discovery workshops with a data-driven baseline that boards and audit committees can trust.

Benefit 3

Migrate without the original vendor

Rewrite your application 1:1 in any target technology — Oracle APEX, Mendix, custom stack — without relying on the original development team or their source code. For UK organisations in regulated sectors where institutional knowledge has walked out the door, this eliminates the most dangerous dependency in any modernisation project.

Four powerful modules

Module 01 — Documentation Generator

Auto-generate complete system documentation — from real usage, not source code.

Traditional documentation efforts rely on developers reading code, interviewing users, and manually writing specs — a process that is slow, inconsistent, and immediately begins going stale. OmniSense observes how the system is actually used and generates documentation from that ground truth.

Research published in the World Journal of Advanced Engineering Technology and Sciences (2025) shows that AI-generated documentation tools can recognise design patterns embedded in legacy code, map data flows between components, and detect business logic that would otherwise require extensive manual reverse-engineering. Critically, 72.4% of developers consider documentation of program functionality essential for understanding existing code — yet most legacy systems have none that is accurate.

For UK-regulated organisations, accurate system documentation is not optional. DORA requires financial services firms to maintain up-to-date ICT asset inventories and system maps. FCA operational resilience requirements demand documented evidence of important business service dependencies. OmniSense generates this documentation as a byproduct of normal system operation.

What OmniSense generates:

#01

Business documentation

Describes what the system does in plain language: workflows, user roles, business rules, and decision logic. Written for stakeholders and product owners, not developers.

Functional documentation

Covers individual screens, form fields, validation rules, and the relationships between system areas. Useful for onboarding, training, and support teams.

Technical documentation

Details component structure, event bindings, DOM interactions, and interface contracts. Enables developers to maintain and extend the system safely — even without access to source code.

Process flows and BPMN diagrams

Automatically maps sequences of user actions into structured process diagrams. Identifies all known branches, decision points, and alternative paths — including ones not covered in any existing specification.

UX analysis and improvement recommendations

Flags usability issues, friction points, and inconsistencies detected during session recording. Includes AI-generated recommendations prioritised by frequency of user impact.

AI opportunity analysis

Identifies screens and workflows where AI automation or AI-assisted interactions would add meaningful value — giving your product and technology teams a concrete modernisation roadmap backed by real usage data.

Coverage gap detection

OmniSense continuously tracks which process paths have been observed and which remain uncovered. It proactively surfaces gaps so they can be explicitly recorded or marked as intentionally out of scope.

As-is vs. existing documentation comparison

If you have legacy documentation, OmniSense compares its generated output against what you already have — highlighting contradictions, outdated sections, and gaps. This alone can save weeks of manual audit work.

Module 02 — In-system AI assistant

A conversational AI built into your application — powered by OmniSense knowledge.

The assistant widget is deployed directly inside your existing system. Because it is backed by the OmniSense knowledge base, it understands the system the way a veteran user does — not generically, but specifically for your application, your terminology, and your workflows.

Capabilities:

#02

Contextual "how does this work?" answers

Users can ask about any screen, field, or feature in natural language and receive an accurate, context-aware response. No more hunting through outdated PDFs or waiting for a colleague.

"What can I do from here?" navigation guidance

The assistant understands the user's current position in the system and can suggest next steps, alternative paths, or related features — reducing errors and shortening task completion time.

"Why am I seeing this error?" diagnostics

When users encounter problems, the assistant cross-references the observed behaviour against known patterns in the OmniSense knowledge base, identifying likely causes before a ticket is raised.

Intelligent bug reporting assistant

The assistant validates whether a reported issue is a genuine system defect or user error, enriches the bug report with session context automatically, and can submit it to your ticketing system without manual effort.

Change request (CR) drafting

Users and business analysts can describe a desired change in plain language. The assistant structures the request into a formal CR with affected screens, impacted flows, and suggested scope — dramatically reducing the time from idea to specification.

Source code independence

OmniSense does not need access to your repository because it generates documentation from observed user behaviour, not by analysing the underlying code.

Module 03 — backoffice tools

AI-powered operations tooling — built on real system behaviour, not assumptions.

OmniSense transforms session data into a set of intelligent tools your teams can use every day, without waiting for a full migration or rewrite.

Regression test generation

One of the most persistent problems in legacy maintenance is the absence of automated tests. Research shows that cross-module dependencies exist in approximately 78% of legacy applications, making manual regression testing both expensive and incomplete.

OmniSense generates regression test suites directly from observed user sessions. Each recorded flow becomes a reproducible test case. The system uses machine learning to identify edge cases and critical paths most likely to break during future changes — ensuring coverage precisely where risk is highest.

Integration test scenarios – derived from multi-screen user flows.
Edge case detection via ML-identified gaps in critical paths
Regression suite generation – behavioural snapshots that prevent functionality loss during modernisation.
Continuous test update – as new sessions are recorded, the test suite grows automatically

#03

Automatic documentation synchronisation

Documentation generated by OmniSense is not a one-time deliverable. As the system evolves and new sessions are captured, OmniSense detects changes in behaviour and updates the relevant documentation sections. Your knowledge base stays current without manual maintenance.

WCAG and compliance checking

OmniSense analyses screen recordings against WCAG 2.1/2.2 accessibility guidelines, flagging violations such as missing labels, poor colour contrast, keyboard navigation issues, and inaccessible interactive elements. Reports are generated per screen and prioritised by severity. For UK public sector organisations and any business subject to the Equality Act 2010, WCAG compliance is not optional — OmniSense provides the evidence base for a structured remediation backlog.

AI feature opportunity detection (AI Proposal Generator)

OmniSense identifies specific screens and workflows where AI tooling — intelligent auto-fill, predictive suggestions, anomaly alerts, or process automation — would produce measurable value. Each opportunity includes a description of the proposed AI feature, the affected workflow, and an estimated impact level based on usage frequency.

Unused module detection

By correlating session data across all users and time periods, OmniSense builds a usage map of the entire system. Modules, screens, and features that receive zero or near-zero traffic are flagged as candidates for removal. Research indicates that approximately 80% of technical debt impact derives from only 20% of the codebase — meaning targeted cleanup of unused and high-risk areas delivers outsized improvements with minimal effort. For UK organisations under DORA, reducing unused system components also reduces the ICT attack surface that must be managed and reported.

Usage heatmap and analytics

Visualise which parts of the system are used most, by whom, and at what times. Identify bottlenecks, high-friction workflows, and underutilised features. Use this data to prioritise modernisation efforts and UX improvements with objective evidence rather than stakeholder opinions.

Module 04 — legacy migrator

Scope, estimate, generate, and validate your new system — grounded in real behaviour.

The Legacy Migrator module transforms OmniSense data into a complete migration package: from scope definition and cost estimation through to AI-assisted generation of the new application, backed by a human-controlled review process at every step.

McKinsey reports that AI-augmented modernisation can accelerate timelines by 40–50% and cut technical-debt-related costs by 40% compared to manual methods — while improving output quality. For UK organisations navigating budget cycles, procurement governance, and regulatory transition requirements, this combination of speed, accuracy, and evidence is transformative.

Feature 1

Scope definition from real usage data

Before a single line of new code is written, OmniSense produces a complete, evidence-based scope document for the new system.

Full system inventory – every screen, module, user role, and workflow observed during data collection.
Dead code and unused feature identification distinguishes active business logic from abandoned functionality, so the new system is not burdened with legacy bloat.
Dependency mapping – traces how system areas interact with each other, with databases, and with external integrations; produces architecture diagrams automatically.
Risk scoring per component – evaluates each part of the system based on complexity, coupling, observed error rates, and change frequency, helping teams prioritise where human review is most important.
Business rule extraction – documents the logic embedded in form behaviour, validation rules, and conditional flows in plain language.

This is equivalent to what AI tools like GitHub Copilot and Azure Migrate now do for code — but applied to the behavioural layer of your web application, regardless of whether source code is available.

Feature 2

Time, cost, and effort estimation

With a complete behavioural scope defined, OmniSense generates a detailed migration estimate.

Screen and workflow count – the raw scope of what needs to be built.
Complexity classification each screen and flow is rated by estimated rebuild effort (low / medium / high / complex).
Technology-specific effort estimates – adjusts estimates based on the target stack (Oracle APEX, Mendix, React + Node.js, custom).
Team composition recommendations – suggested roles, seniority levels, and team sizes based on scope.
Timeline ranges – best-case, expected, and worst-case delivery windows with assumptions stated explicitly.
Cost modelling – produces a structured cost breakdown usable in vendor RFPs, internal budget requests, or UK public sector procurement submissions.

This replaces weeks of scoping workshops and discovery calls with a data-driven baseline that both business and technical stakeholders can trust.

Feature 3

AI-assisted code generation pipeline

OmniSense transforms the behavioural scope into structured prompts and generation pipelines for the new application.

Screen-level generation prompts – each observed screen is translated into a detailed specification that can be fed directly into AI code generation tools or development teams.
Business logic specifications – form validations, conditional flows, and process rules are described in a format that minimises interpretation errors during development.
API interface generation – where the legacy system exposes data, OmniSense generates OpenAPI specifications for modern RESTful or GraphQL equivalents.
Component reuse detection – identifies patterns that appear across multiple screens and flags them as candidates for shared components in the new system.
Data model mapping – maps legacy flat-file or undocumented database structures to modern relational or document-based schemas, with migration scripts where applicable.

Real-world results confirm the potential: Salesforce engineers used AI-assisted migration to complete a rewrite estimated at two years manually in approximately four months. Arcesium reduced API migration effort from 500 person-days to 100 — saving the equivalent of six engineering months on a single project.

Feature 4

Test suite generation before migration begins

One of the most dangerous aspects of legacy migration is losing behavioural fidelity — the new system looks right but behaves differently in edge cases. OmniSense addresses this by generating a comprehensive test suite from observed sessions before the new system is built, giving teams a behavioural contract to validate against.

Unit test generation – derived from individual component behaviours and validation rules.
Regression test suite – full coverage of all observed user flows, ensuring the new system passes every scenario the old one handled.
Integration test scenarios – multi-step flows that cross module boundaries, covering the cross-module dependencies that exist in approximately 78% of legacy applications.
Edge case detection – ML-identified paths that are infrequent but business-critical, such as end-of-month processes or role-specific workflows.
Golden file baselines – sample outputs from observed sessions that the new system must reproduce exactly, enabling deterministic comparison testing.

Feature 5

Validated delivery against the original system

The final phase of the Legacy Migrator is parallel validation: running both the original and new system against the same test suite to confirm behavioural parity before cutover.

Behavioural comparison testing – side-by-side execution of test cases against both systems, flagging any divergence.
Regression gate – the new system must pass all generated regression tests before production deployment is authorised.
Coverage reporting – full traceability from original observed behaviour through to test coverage in the new system.
Rollback readiness – the original system remains operational until validation is complete, enabling safe cutover with a documented fallback plan.
Post-launch monitoring guidance – recommendations for KPIs to track in the weeks following go-live, including performance baselines derived from OmniSense session data.

How it works — from installation to insight

OmniSense follows a structured, milestone-driven process to ensure data quality and documentation accuracy before delivering your final output.

Step 1

PoC preparation (1–2 weeks)

OmniSense installation (at no cost), scope definition, security configuration, anonymisation setup, format selection, and validation of event capture. Data stays within your infrastructure from the first day.

Step 2

Ready to explore (milestone)

OmniSense identifies events, records and correlates sessions, and anonymises data on an environment equivalent to production. Data is anonymised without disrupting business operations. Pretius validates the data and demonstrates the tool's value — you see what OmniSense knows about your system before committing to anything further.

Step 3

Full implementation

Based on the validated PoC, we expand the scope to include full system documentation, an AI agent supporting users, and/or legacy migration tools. You decide what you need. Functionalities are tuned to your system's specific needs, with support from the Pretius UK team at every stage.

Step 4

Documentation (iterative)

Partial docs are generated, reviewed by Pretius and the client in cycles, until full quality and scope are achieved. Documentation does not become stale — it updates automatically as new sessions are captured.

What you get from day one:

No upfront cost beyond compute resources.

Data never leaves your infrastructure.

Works with any web-based system.

Pretius can operate without source code access.

Modular — buy only what you need.

Backed by Pretius AI and legacy migration expertise.

All work governed by English Law through Pretius Ltd.

Why OmniSense matters for UK-regulated organisations

For British enterprises in financial services, insurance, public sector, and other regulated industries, the case for OmniSense is not just operational — it is regulatory.

DORA compliance

– requires financial services organisations to maintain accurate ICT asset inventories, document system dependencies, and demonstrate that important business services can be recovered within tested RTO/RPO targets. OmniSense generates the system documentation and dependency maps that underpin these requirements as a byproduct of normal operation — not as a separate, expensive manual exercise.

FCA operational resilience

– requires firms to identify important business services, map the people, processes, technology, facilities, and information that support them, and prove they can remain within impact tolerances during severe but plausible disruption. OmniSense's process flow documentation and usage mapping directly supports this analysis.

UK GDPR data minimisation and purpose limitation

– for organisations using OmniSense, all session data is anonymised at the point of capture and remains within the organisation's own infrastructure. Pretius does not process personal data in the delivery of OmniSense. For UK DPOs and compliance teams, this architecture eliminates the need for complex data sharing agreements or cross-border transfer safeguards.

WCAG and the Equality Act 2010

– OmniSense's built-in WCAG 2.1/2.2 compliance checking provides the evidence base UK public sector organisations and regulated businesses need to demonstrate accessibility compliance and manage remediation in a structured, auditable way.

The Pretius UK advantage

Local UK accountability

Pretius Ltd is a fully registered UK entity (Company No: 147557660), governed by English Law. Oracle ACE Director Matt Mulvaney leads UK engagements from London. All contracts are UK contracts. All work is 100% IR35-compliant.

Data sovereignty by design

OmniSense is architected so that all captured session data remains within your infrastructure. Pretius can implement and operate OmniSense without ever processing your sensitive business data — satisfying UK GDPR data minimisation requirements and meeting the data residency expectations of UK-regulated sectors.

Instant enterprise scale

300+ in-house salaried specialists. Senior teams deployable within two weeks. No recruitment delays, no IR35 exposure. The same experts who implement OmniSense also deliver Oracle Forms migrations, APEX development, OCI infrastructure, and the full Pretius legacy modernisation service — meaning OmniSense insights can be acted on immediately, by the same team, in the same engagement.

Pretius is the migration partner, not just the tool vendor

OmniSense is built by the same team that has delivered 1,000+ modernisation projects — including the AI Forms to APEX Assistant, which achieves 90% accuracy on effort estimation for Oracle Forms migrations. When OmniSense surfaces a migration opportunity, Pretius can execute it. You do not need to start a new vendor selection process.

By the Numbers

£0

Upfront cost to install OmniSense

1–2 weeks

Time to first documentation output

Average age of enterprise legacy applications targeted

Share of Fortune 500 software built over 20 years ago

40–50%

Acceleration in modernisation timelines when AI is applied

Developers who consider accurate documentation essential

Legacy applications with cross-module dependencies

Technical debt impact concentrated in 20% of the codebase

Pretius in-house specialists available to act on OmniSense findings

ISO 27001 certified · UK GDPR compliant · DORA ready · IR35-compliant engagement

Ready to understand your system?

Let us install OmniSense on your application and show you what it knows in two weeks.

Book a discovery call

Learn about Pretius UK

No upfront cost · Data stays on your infrastructure · Works with any web-based system