Commodities Trading Week Reflections: Execution, Not Experimentation, Is Becoming the Differentiator

Most organisations understand where the industry is heading. They understand the potential of AI, the importance of data, the need to modernise technology estates and the opportunities created by automation. The challenge is no longer one of awareness. The challenge is turning those ambitions into operational reality. 

This shift in focus was evident throughout the event. Compared with previous years, the conversation felt noticeably more mature. AI remained central to almost every discussion, but there was less emphasis on future possibilities and more attention given to practical implementation. Technology leaders are increasingly asking difficult questions about where value is genuinely being realised, how success should be measured, what foundations need to be in place before investment can scale, and why so many initiatives struggle to move beyond isolated use cases. 

AI in Commodity Trading: From Potential to Practical Application 

Perhaps the most interesting observation was that AI is proving to be as effective at exposing weaknesses as it is at creating efficiencies. Organisations that have spent years accumulating fragmented data, complex workflows and disconnected systems are finding that AI shines a very bright light on those problems. 

Faster software development, for example, often reveals bottlenecks elsewhere in the delivery process. Code can be generated more quickly than ever before, but it still needs to be reviewed, tested, approved and deployed. Similarly, automating operational processes frequently highlights issues around data quality, ownership, governance and process design that previously remained hidden beneath layers of manual intervention. 

In many cases, AI is not creating entirely new challenges. It is simply making existing ones impossible to ignore. 

Why Data Strategy and Enterprise Context Matter More Than Ever 

Unsurprisingly, data remained at the centre of the conversation, but the discussion has evolved beyond the familiar message that organisations need better data. 

What emerged repeatedly throughout the event was the idea that data alone is no longer sufficient. If organisations want to move beyond isolated AI use cases and towards genuinely intelligent operations, they need to capture and structure the context that sits around their data. 

Commodity trading decisions are rarely driven by data in isolation. They are shaped by business rules, historical decisions, risk appetite, market conditions, approval processes and institutional knowledge accumulated over many years. Several speakers referred to this as a context layer, enterprise knowledge graph or context graph, but the principle was consistent. 

The organisations most likely to realise value from AI will be those that can combine high-quality data with a deeper understanding of how decisions are made and why they are made. Without that context, AI can generate answers, but it cannot reliably generate judgement. 

Legacy Modernisation Remains a Strategic Priority 

Legacy technology emerged as another dominant theme and, judging by audience feedback, remains one of the most pressing concerns for technology leaders across the sector. 

While legacy platforms have always represented a cost and operational challenge, the discussion highlighted how they are increasingly becoming a strategic constraint. Commodity markets continue to become faster, more interconnected and more data-intensive. Trading hours have expanded, data volumes have grown exponentially and expectations around speed of decision-making continue to increase. 

Against that backdrop, technology estates that were designed for a very different operating environment are struggling to keep pace. 

What was particularly interesting was the suggestion that the real issue is not necessarily modernisation itself, but the order in which organisations pursue it. There was a strong argument that many transformation programmes focus on symptoms rather than root causes. Applying AI to inefficient processes, modernising applications without addressing fragmented data, or accelerating delivery without improving governance can simply create new forms of complexity. 

The organisations making the most progress appear to be those taking a more disciplined approach, identifying the true constraints on performance before deciding where to invest. 

Operational Efficiency Is Emerging as the Strongest AI Use Case 

This focus on practical outcomes also shaped much of the discussion around AI investment and return on investment. 

There was broad agreement that measuring AI value remains difficult, particularly where use cases are exploratory or decision-support oriented. However, there was considerably more confidence when discussing workflow transformation. 

Post-trade processing, reconciliation, trade confirmation, document management and operational monitoring were repeatedly cited as areas where measurable value is already being achieved. These are not necessarily the most eye-catching applications of AI, but they address real operational friction and produce outcomes that can be quantified in terms of speed, accuracy, cost, and efficiency. 

The lesson seemed clear: organisations looking for evidence of value should focus less on grand transformation narratives and more on solving specific operational problems. 

The Human Dimension of Digital Transformation 

Alongside the technology discussion sat an equally important conversation about people. 

Several speakers made the point that AI readiness should not be viewed purely through a technical lens. Data platforms, cloud infrastructure and modern architectures matter, but organisational readiness may ultimately prove more important. 

Successful adoption requires people to understand how their roles are changing, how decisions are being supported, where accountability sits and how new technologies fit into existing operating models. This is particularly relevant in commodity trading environments where expertise, judgement, and experience continue to play a critical role. 

The future is unlikely to be defined by fully autonomous systems replacing people. It is far more likely to involve closer collaboration between human expertise and increasingly sophisticated technology. Creating the conditions for that collaboration requires investment in skills, governance, leadership and organisational change as much as it does in software. 

What Commodity Technology Leaders Should Focus on Next

What became increasingly apparent throughout the event was that many of the challenges facing technology leaders today are not new. Data quality, legacy systems, workflow inefficiencies, organisational silos and change management have all featured prominently in transformation programmes for years. 

What has changed is the urgency with which they now need to be addressed. AI has raised the stakes because its effectiveness is directly linked to the quality of the environment in which it operates. Organisations with strong foundations will be able to move quickly and capture value. Those with fragmented architectures, poor data quality and disconnected operating models may find that AI simply magnifies existing limitations. 

Viewed through that lens, the most important message from Commodities Trading Week was not really about AI at all. It was about readiness. 

The organisations that succeed over the next few years are unlikely to be those making the boldest claims about transformation. They will be the ones doing the harder work of modernising legacy estates, creating trusted data foundations, capturing organisational knowledge, simplifying workflows and building cultures that can adapt to continuous change. 

AI will undoubtedly play a central role in that future, but its impact will be determined largely by the strength of the foundations beneath it. 

For commodity trading organisations, the question is becoming less about whether to invest in AI and more about whether the business is ready to unlock its full potential. That feels like a far more valuable conversation to be having. 

Interested in continuing the conversation? Get in contact here.