Untangling modernisation complexities through strategic collaboration
A leading global U.S. based retail insurer recently asked us for help. Their core services rely on a mix of systems and technologies that evolved gradually, shaped by shifting needs and day-to-day business demands.
The result is an ecosystem that still earns money every minute, but every market shift demands adaptations that push its legacy structure way beyond its original limits.
Does this sound familiar?
From our experience, finance, insurance, legal, large-scale retail, and other compliance-heavy industries – they all share that common pattern. Over time, business needs, regulations, and delivery pressures create environments where systems, teams and processes become deeply interdependent. What looks like one application, is often a tangled web of legacy code, manual workarounds, and fragmented ownership.
In these contexts, complex application modernisation means working across multiple layers – technical, organisational, and operational. It involves changing architecture, workflows and delivery models, while maintaining the systems that simply cannot stop operating.
This is where the true test of competence begins, particularly for an external modernisation partner. Someone who can cut through layers of assumptions and undocumented dependencies and ask the questions that internal teams have learned not to ask.
In long-running systems, complexities become normal, and workarounds blend into processes. And this ability to challenge the status quo is what sets a trusted partner apart from just another delivery vendor.
Is it really all about complexity?
In practice, complexity acts less like a technical barrier and more like a pressure point that reveals strategic weaknesses. It exposes the gap between those who get blocked and those who move forward.
Take the same retail insurer.
When we first stepped in, their operations centre was flooded by 45,000 individual alerts from their systems, which triggered hundreds of daily notifications. This led to millions of tickets pushed to IT teams across different communication paths.
Years of bolt-on integrations and siloed ownership had left the organisation flying blind; troubleshooting was often guesswork, and the cost of guessing wrong showed up in lost quotes, missed renewals and bruised brand trust.
We knew the noise wasn’t a tooling issue, but as a symptom of deeper fragmentation across teams and systems. Instead of chasing alerts in isolation, we mapped how signals moved through the stack and across organisational boundaries. The outcome was a unified telemetry view in Datadog, where every resource could finally be tagged with ownership metadata that had never existed in one place.
Stay competitive and ensure long-term business success by modernising your applications. With our approach, you can start seeing real value even within the first 4 weeks.
With clean insights in hand, we rebuilt the AIOps configuration from scratch:
- dynamic incident models that correlate across cloud, mainframe and on-prem apps,
- routing policies that push only actionable incidents to the right squad, complete with business impact metadata.
The result? Alert volume crashed by almost 80% in just four weeks, and the alerts that remained were bundled into high-fidelity incidents, driving over 60% compression inside the AIOps platform itself.
But the bigger win sits above the telemetry layer. Freed from firefighting, product teams resumed feature delivery; quote latency dropped, straight-through-processing rates crept up, and the insurer clawed back their ground in the hyper-competitive retail market. The improved signal-to-noise ratio increased individual engineer efficiency, enabling a leaner Operations team structure, responding faster to market demands, and cutting related costs by 10%.
None of it happened by chance. It followed the same pattern we’ve applied across regulated sectors: treat observability as an architecture concern, pair it with automation, and let significant data only set the pace of change.
We’ve seen these dynamics play out time and again, not just here, but across other modernisation efforts, revealing recurring dimensions that tend to shape how complexity builds.
True modernisation challenges a reliable partner must be ready to handle
From our work across regulated industries, we’ve identified 13 recurring dimensions that tend to shape the scope and complexity of modernisation projects.
Not all of them need to be present at once for modernisation to become complex. But even a few, when combined with time pressure, interdependencies or regulatory constraints, can turn a focused upgrade into a deeply entangled transformation effort.
Legacy systems integration
Many enterprise systems were never designed for extensibility. Their monolithic architectures tightly coupled components, and years of workaround fixes make integration with modern platforms inherently risky.
Adding new features or services often requires reverse engineering, building custom APIs or rewriting parts of the system just to expose critical functions.
The challenge lies in understanding both the legacy environment and the modern landscape. The most demanding part? Connecting the two without breaking business continuity.
- Monolithic systems are fragile by design: even small changes can cause unintended regressions.
- Tightly coupled dependencies often require custom wrappers or substantial refactoring to enable API-based interoperability.
Data migration and quality
Large-scale data migrations are rarely straightforward, especially when dealing with petabytes of unstructured data, scattered across ageing platforms.
Poorly documented schemas, inconsistent formats, and missing metadata often turn simple transfers into full-scale engineering efforts. Without robust ETL pipelines and validation layers, even minor errors can cascade into major data loss or corruption. And when systems can’t be taken offline, every step must be executed with precision.
- Moving large volumes of data without downtime demands resilient pipelines and robust rollback mechanisms.
- Legacy formats, nested structures and encryption standards may require deep pre-processing before ingestion.
Technical debt remediation
Legacy systems often carry years of accumulated quick fixes, ad hoc patches, and undocumented changes. What may seem like a stable application can hide fragile dependencies, old libraries or conflicting logic layers, revealed only when systems are under load or under change.
- Undocumented code and years of patching make refactoring a delicate process. Teams must weigh the risk of rewriting against the cost of maintaining brittle structures.
- Technical debt impacts planning, delivery speed, and system reliability. Addressing it demands deep understanding of historical decisions and architectural intent.
Multi-cloud transformation
Shifting to cloud-native services is rarely a lift-and-shift operation. It requires rethinking architecture, security, scalability, and operational models.
- Monoliths may require rethinking. Where justified, breaking them into microservices can unlock scalability and speed. But that demands containerisation, orchestration, and rearchitecting, often while still supporting legacy operations that can’t be paused.
- Scaling into cloud also means designing elastic infrastructure by right-sizing resources, tuning auto-scaling policies, and avoiding over-provisioning. Default configurations, especially those provided by cloud vendors, rarely meet production-grade requirements . Achieving cloud-native maturity requires validating vendor recommendations, adjusting accordingly, and testing how services behave under real workloads and constraints.
Security and compliance
While the core security principles, like zero-trust, encryption, and role-based access, are well understood, applying them consistently across organisations is anything but simple, especially when data is processed across legacy systems, cloud-native platforms, and external parties, each subject to different controls and compliance expectations.
- Enforcing identity and access management across cloud and on-prem workloads often requires bridging incompatible systems.
- Coordinating secrets management, network segmentation and audit trails at scale demands both technical integration and shared governance.
- Meeting compliance requirements (e.g. GDPR, HIPAA, sector-specific localisation rules) may delay releases or trigger costly rework when not planned from day one, and remain a moving target, as sudden regulatory changes may disrupt even well-structured delivery plans.
- Limiting access for third parties without valid data processing agreements requires enforceable boundaries, full auditability, and in-transit encryption—especially when queries run on production datasets.
API Management & Interoperability
Legacy systems rarely come with clean, documented interfaces. Most modernisation projects involve wrapping core functionality in new APIs or rebuilding integrations that have grown organically over the years. Does building APIs sound straightforward enough?
Well, the real challenge is managing them. As legacy and cloud-native systems co-exist, versioning and compatibility must be carefully handled to avoid breaking the existing integrations. With growing traffic and multiple external users, maintaining availability, uptime, and performance becomes a challenge on its own.
- API versioning strategies are required to maintain backward compatibility while evolving feature sets.
- As integration complexity increases, SLAs, rate limiting, circuit breakers, and observability become critical to avoid cascading failures.
Organisational & cultural change
Modernisation cuts across the status quo, exposing misalignments between IT and business, between teams working in silos, or between current capabilities and future goals.
In this environment, a modernisation partner must take the lead. They must reduce resistance, drive adoption, and prove, early on, that change is worth the effort.
- CI/CD, observability, and automation won’t stick unless they deliver visible gains in speed or efficiency of delivery teams.
- Upskilling and process alignment require structure and ownership.
- Without guidance and accountability, teams fall back on legacy habits, making modernisation stall.
- Measurable results, such as faster release cycles or reduced incident load, are what ultimately shift mindset and unlock long-term momentum.
Governance and architecture standards
Architectural governance is a way of keeping strategy and execution aligned, even as priorities shift. Whether it’s a choice of frameworks, cloud-native design principles, or how teams handle security or cost-efficiency, it’s always the standards that shape the outcomes.
- Architecture standards provide a shared baseline for change across teams, vendors and environments.
- Governance practices tie implementation to business outcomes, tracking both velocity and value.
Without clear direction, decentralised efforts fragment; with it, teams can move fast and still stay aligned.
Continuous delivery & testing
Frequent releases are only safe if backed by automation. But in legacy-heavy environments, testing is often the weakest link.
We often see test coverage below 20%, especially in systems that evolved without consistent quality engineering. Although it’s not a blocker, it still means that modernisation needs to start with enabling visibility to know what’s covered, what’s critical, and how failure propagates.
- Test automation enables safe, repeatable releases even under pressure.
- Feature flags, canary deployments and rollback plans reduce the blast radius of change.
- High-confidence pipelines are not built overnight, but they are built deliberately.
User experience and change management
At the end of the day, modernisation always lands on people. And people often resist what disrupts their routines. If change happens too fast, adoption stalls. Too slow, and legacy problems linger, making the users even more frustrated or quit.
That’s why experience design and change enablement must be embedded into the delivery process, not treated as an add-on or an afterthought. Interfaces, behaviours, and habits need space to evolve.
- Gradual rollouts, feature flags and opt-ins help reduce user resistance.
- Real-time feedback from users helps detect friction early and prioritise meaningful fixes.
- Change management means ongoing support, training and listening.
Observability and monitoring
With systems spanning clouds, microservices and teams, observability becomes the backbone of operational readiness.
To understand what’s really happening, teams need structured tracing, centralised logging, and real-time visualisation, ideally integrated into the development and deployment process.
- Correlating logs, traces and metrics enables faster root-cause analysis and reduces mean time to recovery.
- Without proper tuning, alerts quickly become noise. Intelligent thresholds and routing logic help teams focus on what matters.
- Observability makes complex systems understandable (and recoverable) at scale.
Cost & ROI optimisation
While not every modernisation starts with cost in mind, nearly all of them touch it eventually.
Infrastructure, licensing, delivery, and support models often need to be reassessed: to justify the investment and to avoid replicating old inefficiencies in new environments.
- Cost reviews during modernisation often reveal where architecture, delivery and business goals are out of sync, creating opportunities for structural efficiency gains.
- Disciplines like FinOps bring accountability by tracking granular usage, tagging spend, and enforcing budget policies across teams.
- Balance is key: cost-efficiency must be weighed against resilience, speed and business impact.
Read more: How does infrastructure modernisation help reduce IT costs?
AI readiness and integration
A common misconception around AI adoption is that it starts with choosing the right tools. But asking “what tools should we implement to become a fully AI-driven organisation?” is the wrong question. And most often a sign that critical groundwork hasn’t been done yet.
Successful AI integration relies on a coherent ecosystem: unified data structures, clean flows, defined ownership, and consistent governance. In reality, AI amplifies whatever foundation it’s been given. This means that without modernisation, it often multiplies legacy issues rather than unlocking value.
Achieving AI readiness requires the same systematic thinking that drives effective modernisation:
- Clean architecture and structured data, so models are trained on noise or fragmented, unreliable input.
- Strong governance and boundaries, to ensure safe, compliant, and explainable outcomes.
- Clear service ownership and flows, so AI can augment (and not confuse) decision-making
- Platform-wide consistency, enabling traceability, performance monitoring, and feedback lops.
These factors form the reality most modernisation projects must navigate, especially in sectors when the cost of failure is high and the room for error is small.
Modernisation with no pause button
Among the many cases we’ve encountered, one that captures the full scope of these challenges comes from a British insurer operating in the London Market – a scenario that reflects patterns we’ve seen across many industries and regions.
At the centre of their operations sits a twenty-year-old core platform, originally built for a single purpose, but now burdened with nearly fifty integration points and business-critical responsibilities it was never designed for.
The insurer has invited us to join the discussions around the future of the system, as part of a broader digital overhaul across the London Market ecosystem. Any major shift could have significant operational implications. And without a clear strategy, the organisation risks losing alignment with the evolving market landscape.
The platform supports essential business operations, including underwriting and regulatory reporting. Over time, these functions have become deeply embedded in day-to-day workflows to the extent that no single team holds a complete picture of how the system operates.
Documentation is limited, and key integration paths have evolved organically, without consistent central oversight.
What remains is a critical engine that must continue to run.
Modernising this kind of platform means carefully mapping ownership, clarifying undocumented flows, and rebuilding ad connections, all while keeping operations fully active. Even short interruptions could disrupt critical processes and downstream partner systems.
Out of the thirteen modernisation dimensions we described earlier, this single system activates nearly all: undocumented legacy, fragile integrations, critical data flows, compliance pressures, user experience, and architectural rethinking. The complexity is embedded in how the organisation trades, reports, and collaborates every day.
This is not a challenge that can be solved with tooling alone. It requires a partner who can operate across technical, operational, and organisational layers, and who understands the business domain well enough to align change with commercial and regulatory realities.
Someone who can help the organisation navigate uncertainty, maintain stability, and steadily move towards a modern, sustainable core. That’s the role we’ve stepped into.
Modernisation in complex environments exposes the full range of what’s required to lead the change
What sets a successful modernisation apart is the ability to recognise where the real complexity lies, and act accordingly.
It highlights the value of a partner who can navigate constraints without losing sight of the bigger picture. Where legacy systems, regulation and daily operations collide, success depends on the ability to map interdependencies, make decisions grounded in context, and protect what keeps the business running.
The right partner brings visibility across both business performance and delivery capability, using a metrics-driven modernisation approach to steer technical execution and to connect modernisation outcomes with what matters commercially.
Doing that well, quietly, deliberately, and side by side with the organisation, is what defines a modernisation partner for complex challenges.
Assure seamless migration to cloud environments, improve performance, and handle increasing demands efficiently.
Modernisation of legacy systems refer to the process of upgrading or replacing outdated legacy systems to align with contemporary business requirements and technological advances.