For decades, the central question for any firm designing its trading systems architecture has been a seemingly binary choice: buy an off-the-shelf platform or build a proprietary one in-house? The ‘buy’ camp argued for speed to market and vendor-managed upkeep, while the ‘build’ camp championed bespoke functionality and control over intellectual property.
Today, this long-standing debate is becoming obsolete. A new, more sophisticated paradigm has taken hold – ‘buy AND build’ – a topic of such critical importance that it will be a key theme of discussion at A-Team Group’s upcoming Buy AND Build Summit in London. This hybrid approach, where firms strategically integrate best-of-breed commercial components with proprietary, differentiating code, is no longer just an option, it’s a strategic necessity for survival and growth in modern capital markets.This shift is not arbitrary. It is a direct response to a perfect storm of technological advancement, intense regulatory scrutiny, and acute budgetary pressures. Firms that cling to monolithic architectures, whether bought or built, risk being outmanoeuvred by more agile competitors. The future belongs to those who can master the art of deconstruction and composition, creating a trading stack that is both resilient and responsive.
The New Calculus: Why Monoliths Are Failing
Traditional monolithic systems, once prized for their perceived stability and single point of accountability, are now showing their age. Their tightly-coupled architecture makes them notoriously difficult and expensive to change. A simple update to one function can require a full regression test of the entire system, leading to glacial release cycles that are fundamentally at odds with the pace of modern markets. This inherent inflexibility is being critically exposed by three powerful forces.
First, regulatory drivers like the EU’s Digital Operational Resilience Act (DORA) are changing the game. DORA demands that firms possess a granular understanding of their technology dependencies, map their critical services, and demonstrate the ability to substitute components to ensure operational resilience. This is a nightmare for firms running on a single, opaque ‘black box’ system from a vendor. A modular architecture, by contrast, aligns perfectly with DORA’s principles. It allows individual components, such as a market data handler, a risk engine, or an order router, to be independently assessed, tested, and, if necessary, replaced without disrupting the entire ecosystem.
Second, the economic imperative is undeniable. In an environment of tightening budgets for technology and innovation, firms can no longer justify the expense of building everything from the ground up, especially the ‘plumbing.’ The focus must be on directing precious capital and engineering talent towards what truly creates a competitive edge, be it a unique execution algorithm, a sophisticated pre-trade analytics tool, or a superior client-facing portal. By buying commodity components, firms can delegate the undifferentiated heavy lifting and concentrate their resources on building true value.
Finally, technological enablement has made this modular vision a practical reality. The rise of cloud-native computing, robust API standards, and microservices architecture has provided the tools to dismantle the monolith and reassemble it as a more flexible, powerful, and distributed system.
The Blueprint for an Agile Trading Stack
Building agility into the trading stack begins with a clear strategic principle: differentiate or delegate. For every function within the trading lifecycle, firms must ask a simple question: is this a commodity service, or is it a core source of our competitive advantage? The answer dictates the approach. Buy the commodity, build the differentiator.
Executing this strategy depends on a set of modern architectural pillars that enable seamless integration and communication between bought and built components.
APIs as the Universal Glue: Application Programming Interfaces (APIs) are the most critical element, serving as the contract-based interface between disparate services. The industry is moving beyond legacy FIX connections towards modern, lightweight RESTful APIs and, increasingly, event-driven architectures. This allows a commercial OMS to communicate flawlessly with a custom-built smart order router, or an in-house algo to consume data from a third-party analytics engine, for example.
The Cloud Foundation: The cloud provides the elastic, scalable, and resilient infrastructure required to host a distributed trading system. By leveraging Infrastructure-as-a-Service (IaaS) and Platform-as-a-Service (PaaS), firms can provision resources on demand, reducing capital expenditure and operational overhead, and freeing up teams to focus on application logic, not infrastructure management.
Containerisation as the Delivery Mechanism: Technologies like Docker and Kubernetes are essential for packaging and deploying custom-built components. Containers ensure that applications are portable and behave consistently across development, testing, and production environments. Kubernetes orchestrates these containers, allowing services to be scaled independently to meet real-time demand, enabling rapid, reliable, and isolated deployments.
High-Performance Messaging as the Nervous System: To facilitate real-time data redistribution, a high-throughput, low-latency messaging fabric is non-negotiable. Solutions like Apache Kafka, Solace, or Aeron serve as the central nervous system of the trading stack, capable of streaming market data, orders, and execution reports between all modules, both bought and built, with reliability and immense scale.
Strategic Use of Open Source: Firms are increasingly leveraging open-source software to accelerate development of their custom components. By using proven open-source libraries for functions like logging, monitoring, machine learning, or UI frameworks, engineering teams avoid reinventing the wheel and can build their differentiating logic faster.
Beyond Technology: The Cultural and Organisational Shift
Adopting a ‘buy and build’ model is as much a cultural and organisational challenge as it is a technological one. A modular architecture cannot thrive in a siloed organisation.
The role of the central IT department must evolve from being monolithic system owners to becoming expert service integrators and platform enablers. Their mission is to provide the core infrastructure, tools, and guardrails that empower business-aligned teams to build and integrate services safely and efficiently.
This requires breaking down traditional functional silos and embracing a DevOps culture with cross-functional teams. Small, agile ‘squads’ or ‘pods,’ comprising developers, quants, operations, and business experts, should be given end-to-end ownership of a specific business capability, like an FX derivatives pricing service. This organisational structure mirrors the modularity of the architecture, fostering accountability, speed, and innovation.
Finally, this transition demands a mindset shift away from ‘big bang’ projects and towards continuous evolution. The goal is no longer a single, multi-year implementation but a constant cycle of iterating, improving, and deploying small, incremental changes. This requires a culture that embraces experimentation and is comfortable managing a more dynamic, heterogeneous technology landscape.
The ‘buy and build’ approach represents a fundamental maturation of financial technology strategy. It moves beyond simplistic binaries and towards a sophisticated, pragmatic model for building trading systems. The firms that master this hybrid approach, aligning their technology, their processes, and their culture around the principles of modularity and agility, will be the ones who can innovate faster, navigate regulatory change, and ultimately win in the demanding markets of tomorrow.
Subscribe to our newsletter
