For decades, the Order Management System (OMS) and the Execution Management System (EMS) occupied distinct roles within the trading process; the OMS handling the full order lifecycle, from creation to post-trade processing and compliance, and the EMS managing real-time order execution, routing, and optimisation across markets and venues.
Today, a powerful convergence is reshaping this landscape, raising a series of critical questions for the industry. How are the relentless demands of multi-asset trading forcing a rethink of siloed technology? What does the fundamental shift towards open, interoperable platforms and a ‘buy and build’ philosophy mean for firms and their vendors? And beyond the hype, how is the practical application of artificial intelligence truly changing the game for traders?Ultimately, what does this all mean for the future of the trading desk?
The End of an Era? The OMS and EMS Divide
The relevance of the traditional OMS/EMS split now largely depends on which side of the street you are on.
“On the buy-side, we see two very distinct camps taking different approaches,” observes Dan Enstedt, Director Of Sales And Business Development EMEA at Flextrade. “The long-only community generally wants to keep the OMS and EMS segregated. They want a seamless experience, but they value having things fully decoupled because the two systems serve different functions and evolve at a different pace. The EMS needs to be nimble to adopt new broker algos and analytics, which happens much faster than the OMS layer evolves. Then you have the hedge fund side, where the trend is the complete opposite. They want everything in one O/EMS, no exceptions. The big driver is having one consolidated view of risk, especially in multi-strat firms. They need to spin up a new strategy and deploy capital across anything from global macro to merger arb, and be able to monitor that risk instantly from a single platform.”
The picture on the sell-side is a little different. For many larger banks and brokers, integrated O/EMS platforms have become the standard, although rather than a single monolithic system, most are adopting modular platforms where order control, risk, data and entitlements sit on a shared infrastructure, and different execution engines plug in by asset class, to create a converged, interoperable stack where the OMS and EMS act as components of a broader trading platform.Munish Gautam, Global Head of Trading Platforms Product Management at Broadridge, elaborates. “Relentless commission compression and regulatory change is forcing a complete rethink of the trading technology stack, demanding unprecedented levels of efficiency,” he says. “This has accelerated the convergence of trading functions, with desks and asset classes merging into more integrated teams. To support this new reality, firms require a fully integrated, front-to-back platform that can handle multi-asset risk, high-touch, and electronic trading in a single, seamless workflow. The key to achieving this is not a rigid, monolithic system, but rather a flexible platform built on interoperability, using APIs to connect the different technologies needed for order management and low-latency execution into a unified whole.”
Building Depth on a Foundation of Breadth
Underpinning this evolution is a profound shift in how financial institutions approach technology. In place of monolithic solutions, a hybrid ‘buy and build’ model is becoming the industry standard.
“A guiding philosophy for the modern vendor-client relationship is the principle: ‘You buy breadth; you build for depth.’ This reflects the reality that a vendor cannot be a specialist in everything,” points out Gautam. “The vendor’s role is to provide the breadth – a service-based, customisable, ready-to-deploy OMS with various modular components, and an EMS toolkit that is roughly 80% complete out of the box on the execution trading side. The client then builds the depth – the final 20% of deep, specialist logic where they create their unique trading alpha, such as custom algorithms or proprietary market-making models. This entire approach hinges on total interoperability, requiring platforms to be built with fully integrable APIs that allow clients to seamlessly plug in their own applications and analytics.”
This is especially true for the most technologically advanced firms, notes Enstedt.
“The trend is particularly pronounced in the hedge fund space,” he says. “Their alpha generation is their entire business, so they demand the ability to build and control their custom trading logic. To support this, a platform must be built API-first from the ground up. The expectation is that if you can do something in the UI, you can do it via the API. This allows their developers to interact directly with the OMS, integrate their proprietary models, and manage orders programmatically, which is a perfect use case for a modern, open platform.”
This model is entirely predicated on the central role of APIs and interoperability. A platform’s value is now measured not just by its features, but by its openness.
“An open, modular, API-first approach is key,” states Ovidiu Campean, Director of Product, Execution Solutions at LSEG. “It gives clients the flexibility to decide which out-of-the-box modules they use and what they implement regarding their own proprietary models or logic. A large firm, for example, wants to implement its own custom routing rules or algo wheels; they don’t want to be forced into using a vendor’s automation logic. Having an API-first platform is attractive because it allows them to do this – they can implement their own signals, send and manipulate orders, and read positions, all driven by their unique logic.”
From Open Platforms to Productised Infrastructure
A high degree of openness is no longer optional, it is now a core requirement, according to Isabelle Dominjon, Head of OMS Strategy & Sales at Horizon.
“In every discussion we have, two requirements are constant: clients want to connect to their chosen best-of-breed analytics providers, and they want the ability to integrate their own internally-developed analytics into the OMS,” she says. “At Horizon, our entire framework is built to meet this demand. Through APIs, clients can easily integrate external or internal tools directly into the blotter and control any feature of the platform. For firms with strong development teams, this level of openness is no longer a nice-to-have, it’s an absolute necessity.”
She continues: “The ‘buy and build’ approach defines our most successful client partnerships. We provide the core platform with its standard trading and risk services, and then we interface with what our clients are building to help them create their competitive edge. This means that clients can create their own proprietary pricing and strategy models and inject that logic directly into the platform, allowing their unique value to sit on top of a robust foundation.”
Taking this concept a step further, some providers are now productising the very infrastructure that enables this connectivity.
“I often use the analogy of Amazon to explain what’s happening,” explains Medan Gabbay, CEO of Quod Financial. “Amazon built a global online shop, but to do so, they first had to solve the problem of hosting and serving massive amounts of content globally. They built an architectural solution for themselves, and then realised other people could use it – that solution became AWS. We had a similar journey. To build our O/EMS, we first had to create a sophisticated application integration layer – a trade-aware, back-end infrastructure integrated with hundreds of services. If you remove the O/EMS from the picture, what you’re left with is a unified, trade-aware, pre-built technology stack. This layer can connect to any system, normalise the data, and reconstitute orders automatically. We realised that this foundational technology – the part that solves the core integration challenge – is arguably more valuable to the market than another O/EMS replacement.”
The Multi-Asset Mandate: Challenges and Solutions
While many platforms claim to be ‘multi-asset’, traders often still face significant workflow friction.
“The term ‘multi-asset’ is often used too loosely and rarely reflects true all-asset capability; a genuinely multi-asset platform must support both electronically and manually traded instruments, along with full post-trade lifecycle events,” argues Marc Schröter, Chief Product & Technology Officer at SimCorp. “Key challenges include fragmented liquidity in complex instruments, disparate OTC workflows, and the need for a bulletproof underlying data architecture to unify execution, risk, and compliance. A unified platform that addresses these gaps can eliminate silos and deliver a seamless cross-asset experience.”
One of the biggest and most persistent barriers to achieving this is the data normalisation challenge, according to Campean.
“The foundational barrier to effective multi-asset trading is the lack of normalised data,” he says. “If data is not normalised at the source, it becomes incredibly difficult to ingest, forcing the creation of an inefficient and costly translation layer within the system. When a normalised data set is available from the outset, however, it becomes far easier to inject information into a single, unified platform. This is where open APIs become critical, as they allow a system to connect to a wide array of specialised liquidity pools while still presenting the information to the user in a consistent format. For large global macro funds that trade everything, this is invaluable; it allows them to operate from a single system, eliminating the need to switch contexts between different applications, while still accessing the best liquidity for each specific asset class.”
This is especially true when bridging the gap between listed instruments and OTC markets, notes Enstedt.
“While workflows for listed instruments such as equities, futures, and FX have become relatively simple to integrate, the greater challenge lies in bridging the gap to OTC fixed income,” he says. “The fixed income market is inherently more fragmented, with historically segregated workflows that have led to a reliance on specialist, standalone platforms.
Consequently, it is in this asset class that we are now seeing the most significant innovation. The focus has shifted towards building robust APIs and opening up platform capabilities to enable deep, streamlined integration. This moves away from the old paradigm where traders were forced to use separate screens for different platforms, which then had to be manually connected to risk and compliance systems. The ultimate benefit of this evolution is the ability to run compliance, risk, and exposure analysis holistically across all asset classes within a single, unified OEMS.”
AI & the Blurring of High-Touch and Low-Touch
With artificial intelligence moving steadily beyond the hype and starting to have a tangible impact on the trading desk, one of its most immediate effects is the blurring of the lines between high-touch and low-touch trading, observes Holden Sibley, Head of Investment Management & Execution Solutions at LSEG.
“For years, the trading world was sharply bifurcated between high-touch and low-touch workflows. The high-touch process, while allowing for human interaction, was notoriously inefficient, often culminating in the manual, ‘swivel chair’ process of transcribing a deal from a chat window into a booking system. The advent of Natural Language Processing (NLP), now significantly enhanced by Large Language Models (LLMs), is dissolving this divide and seamlessly blends the two worlds, retaining valuable human interaction while injecting the end-to-end efficiency of automation. The result is a ‘smoothing of the curve’ between the two formerly separate domains, giving traders the unprecedented flexibility to move fluidly between high-touch and low-touch approaches depending on the liquidity and context of each specific trade.”
Beyond automation, AI is augmenting the capabilities of human traders, creating new forms of value, by providing both decision support and a new ‘operational alpha’.
“The impact of AI is being felt most immediately in the front office, where nimble, API-driven systems can quickly integrate new tools,” observes Gautam. “We are seeing this in the transformation of TCA from a backward-looking analysis to a real-time, forward-looking decision support tool. AI ‘copilots’ can now continuously analyse market data and provide live recommendations on execution strategy. However, the biggest ‘bang for the buck’ is expected to come from the middle and back office in the long run. This is where AI will deliver operational alpha by solving deep-seated challenges. For instance, instead of assuming settlement fails will always happen and focusing on minimising them, AI can now be used to predict which specific trades are likely to fail, allowing firms to prevent the problem entirely and unlock significant efficiencies.”
Looking further ahead, the very nature of human-computer interaction is set to evolve, says Gabbay.
“Where we’re heading from a platform perspective is fascinating. We have a core logic engine that allows for complex ‘if-then’ rules for routing, order modification, and so on. On top of that, we have a visual builder that translates this logic into scripting. But the most significant step, which we’re working on now, is placing a language model on top of that scripting layer. This will allow a user to simply state in natural language, ‘When an order meets these criteria, route it here and set a flag,’ and the system will generate the code framework to build that automation directly into the platform. It’s a fundamental shift from coding APIs to creating complex business workflows through natural language interaction.”
The Path Forward: An Agile, Integrated, and Intelligent Future
It’s clear that the future of the O/EMS belongs not to monolithic, closed systems, but to open, interoperable, and intelligent platforms. This new paradigm is enabled by the ‘buy and build’ philosophy, where firms leverage the breadth of vendor platforms to build their own deep, proprietary value.
As Marc Schröter of SimCorp concludes, this technological shift will ultimately drive a convergence of human roles as well.
“Technology can actively drive the convergence of investment roles by making unified data – real-time cash, positions, trading activity, and historical insights – accessible across the entire investment lifecycle,” he says. “Platforms that replace siloed modules with workflow-specific components and offer a Copilot overlay empower users to create tailored views, collaborate seamlessly, and leverage shared analytics and AI to enhance decision-making and break down functional barriers.”
Subscribe to our newsletter
