Fei, please tell us a little about what your job involves.
I work in a corporate treasury trading function where I oversee FX execution across a wide range of currency pairs, including spot, forwards, NDFs, and swaps. My role centers on managing market-facing execution and optimizing outcomes through algorithmic strategies, especially for large-value trades. These tools help us reduce market impact, control execution costs, and minimize information leakage.
Beyond execution, I help design and implement strategic hedging strategies aligned with macroeconomic conditions, using data-driven insights to manage currency risk across regions. I also lead automation initiatives to streamline workflows, integrate analytics, and enhance real-time visibility. Collaborating closely with liquidity providers and internal stakeholders, I bridge market expertise with operational precision to support business objectives globally.
What range of instruments are treasurers generally working with?
From what I see in the industry, corporate treasuries primarily work with FX spot, forwards, swaps, and non-deliverable forwards (NDFs). These instruments form the core of hedging programs aimed at managing transactional and translational currency risk.
We also see the use of interest rate swaps and cross-currency swaps to manage funding and balance sheet risk. Additionally, some corporates use basic options strategies – such as collars – to provide downside protection while preserving some upside, particularly when hedging budget rates in volatile environments.
While the instrument set is typically more risk-focused than return-seeking, the sophistication of how these tools are deployed continues to evolve with hedging objectives and market conditions.
How important are FX execution algos becoming in treasury trading activities, and what are the main objectives when using them?
Algos are becoming a core part of FX execution infrastructure for many corporates. Market- wide, algo usage has grown significantly—Bloomberg reports it has risen from about 20% 5 years ago to nearly 50% today. This adoption trend underscores a broad shift toward automation, driven by the need for more efficient and data-informed execution practices.
The main objectives include reducing market impact through intelligent order slicing, improving execution precision across fragmented liquidity pools, avoiding information leakage—especially in less liquid pairs or large trades—and enhancing consistency and scalability, particularly across regions or time zones. Algos are no longer niche tools; they are now foundational for firms that care about execution quality, data, and control.
How would you summarise some of the key benefits of utilising FX execution algos, especially in the corporate treasury environment?
FX algos offer strong advantages for corporate treasuries focused on efficiency, control, and transparency. They reduce market impact by slicing orders intelligently and improve cost outcomes, especially in less liquid pairs. Algos also allow firms to tap into diverse liquidity sources with minimal information leakage.
More importantly, they enhance operational scalability. With pre-trade analytics, real-time visibility, and post-trade TCA, algo execution gives treasury teams the flexibility and insight needed to trade confidently in a fast-moving global market.
What issues need to be considered when it comes to going about sourcing FX execution algos, and what factors might influence the decisions that have to be made here?
Selecting algos should start with aligning the offering to your specific execution objectives. Key considerations include the breadth of strategies available (TWAP, VWAP, liquidity-seeking, passive/aggressive), coverage across currency pairs (especially in restricted markets), and the underlying liquidity sourcing model—whether it’s venue-agnostic or favors internalization.
Another important factor is the integration of pre-trade, in-flight, and post-trade analytics. These tools provide transparency and enable traders to calibrate execution strategies in real time. In-flight analytics, in particular, allow traders to respond to shifts in volatility or spreads mid-trade—a critical feature during turbulent market conditions.
The ability to configure execution parameters, such as speed, execution style, and venue preference, also matters. Strong sales coverage, clear documentation, and regular performance reviews further influence provider selection. Ultimately, decisions should be guided by measurable execution outcomes, not just brand recognition. Firms should pilot different providers and review their slippage, spread capture, and participation rates to identify the most aligned and consistently high-performing partner.
In addition, alignment with the FX Global Code is an increasingly important consideration.
Firms should ensure that their algo providers adhere to the principles of fairness, transparency, and governance outlined in the Code. This reinforces ethical standards across the execution process and builds trust with internal stakeholders and external counterparties alike.
Some traders are happy to let an algo do its work without much oversight, whilst others prefer more real-time visibility during the execution process. What’s your opinion about that?
I believe real-time visibility is critical. Market conditions change rapidly—liquidity can vanish or volatility can spike—and being able to adjust parameters mid-execution is a major advantage.
In-flight TCA and adaptive controls allow traders to manage slippage, execution speed, and venue selection dynamically. Algos are powerful, but they’re most effective when paired with experienced oversight and tactical decision-making.
Analysing the results of algorithmic trading is very important. How can this be done to see how effective it is and whether the strategies are meeting execution objectives?
Post-trade analysis is a cornerstone of FX algo adoption. It provides objective, data-driven feedback on how well a strategy performed relative to expectations. Traders typically evaluate slippage versus arrival price, risk transfer price, TWAP or VWAP benchmarks, participation levels, fill ratios, and overall execution efficiency.
Third-party TCA providers like BestX are often used to provide independent validation and enable benchmarking across providers and time periods. This allows us to assess whether one provider consistently outperforms others in specific currency pairs or under certain market regimes. The insights gained from these reviews help inform future strategy selection, flow allocation, and provider engagement. Over time, this structured analysis ensures that execution remains optimized, transparent, and aligned with business goals.
In what ways is leveraging data and analytics becoming more important in helping traders make more effective use of FX algos?
Data and analytics are central to the full lifecycle of FX execution. Pre-trade tools help select the right strategy based on liquidity and volatility conditions. In-flight analytics provide real-time insights, allowing for course correction during execution. Post-trade TCA completes the loop by validating performance and informing future optimization.
Traders today need to be comfortable with interpreting this data and translating it into action. It’s what makes algo execution more than just automation—it becomes a tool for ongoing improvement.
What about algo wheels? What are your thoughts about the relative merits of using those?
Algo wheels can support fairness in provider allocation and reduce human bias in routing flows. They also simplify performance tracking by standardizing how trades are assigned across banks.
That said, the benefit is marginal if you already allocate flow based on real executionquality. For example, some corporates may allocate algo flow based on actual tradingperformance. This approach allows them to reward providers who deliver the best resultswhile keeping control over how flow is distributed. For some, wheels are helpful; for others, disciplined oversight can be equally effective.
How far do you expect to see next generation technologies like AI makingtheir presence felt in the trading room over the next few years and being used in applications like algorithmic FX trading?
AI is already beginning to influence FX trading, and I expect its role to expand rapidly in both execution and strategy design. One of the most promising applications is in pre-trade decision-making: using AI models to determine whether a trade should be executed via RFQ or algorithm based on expected cost, urgency, and market conditions. I would also expect the use of AI in selecting the most suitable algo strategy based on real-time liquidity, volatility and cost projections – essentially turning strategy selection into a smarter, data- driven process.
During execution, AI can power adaptive algorithms that adjust participation rates, aggression levels, or venue selection in real time, optimizing fill quality while minimizing slippage. Post-trade, AI can be used to detect patterns in slippage or identify anomalies in execution performance that human analysts might miss. While traders won’t be replaced, their toolkit will become increasingly intelligent – freeing them to focus on high-impact decisions rather than manual interventions.
In what ways is the growing use of technology and advanced toolsets like algorithmic trading likely to impact on the future skillsets required of top-class traders like yourself and how will it change the nature of the trading room environment that you are operating in?
The future trader must be as comfortable with data and systems as they are with markets. As execution becomes increasingly electronic, traders are evolving into strategic architects configuring execution flows, analyzing TCA, and managing provider performance.
The trading environment is shifting from manual to strategic. With many low-touch trades automated, the focus is on large orders and execution strategy development. The future trading room will be less about button-clicking and more about shaping execution workflows, managing tools, and adapting in real time.
Do you expect to see more use being made of algorithmic trading toolsets in FX in the future?
Absolutely. As FX markets grow more electronic and fragmented, algos offer scalable, cost- effective execution. Systematic hedge funds already rely heavily on algos, and these capabilities are now spreading across real money and corporate desks. With growing confidence in algo performance, expanding support across EM currencies, and greater integration with analytics platforms, I expect algos to continue gaining share becoming the default execution method in many FX programs.
What advice would you give to firms who may be considering using FX execution algos and looking to exploit the power of these powerful toolsets?
Start by defining clear goals for using FX algos—whether it’s reducing trading costs, improving transparency, or enabling scalability. Firms new to algo execution should begin with a small set of currency pairs and monitor results closely using pre-trade analytics and post-trade TCA.
It’s important to view algo execution as a strategic capability, not just an automation tool. This means investing time in understanding each provider’s strategy logic, backtesting assumptions, and regularly reviewing performance. Traders should also be trained to adjust algo parameters in real time based on changing market conditions.
Partnership is key—collaborate closely with algo providers and independent TCA platforms to ensure continuous improvement. With the right foundation, FX algos can meaningfully enhance execution performance and help institutional desks scale and compete in a more electronic marketplace.
