From Tech to Trade: Building Success in Commodities

From conversations with CEOs and COOs, it’s apparent that companies will often split into two camps: sceptics and optimists. Sceptics believe that technology has too many limitations to have a real impact no matter how much they try. Optimists believe that technology will revolutionise everything for the better with very little effort. The truth lies somewhere in between.

It’s extremely important to balance these two dynamics. A healthy dose of scepticism helps to flush out the generic tools built to tackle generic problems. And that’s a good thing. It saves time and money. 

But as Kahneman explains in “Thinking Fast, Thinking Slow”, we often have a tendency to place more weight on negative information through negativity bias. It sounds more clever to be pessimistic. 

The risk, then, is that scepticism reaches such a level that it prohibits senior management teams. They dwell on the downside and quite often rely on technical people who might not be as strong at spotting commercial opportunities. So they decide to go for “low hanging fruit”, off-the-shelf products or only look for efficiencies. 

As a result, many organisations today aren’t risking enough time and money to leverage technology in transformative ways that will deliver profit. Low-hanging fruit quickly goes off – and inefficiencies plague larger corporations more than nimble businesses. As they say, money often transfers from those who are more pessimistic to those who are more optimistic. 

So how do companies find this balance? There are two ways. 

Firstly, realise that important developments in technology today are driven by fundamentals, not hype. People look at AI or Nvidia or Bitcoin and panic that everyone else has lost their minds. It’s not true. The technology we see and use today is being driven by a very fundamental economic principle: the collapse in the cost of data storage, as argued by economists such as Bram and Schmalz.

When something becomes cheaper, we use more of it. In 2010, there were approximately two zettabytes of data generated worldwide. By 2024, this figure was 149 zettabytes. That is a staggering 7,000% increase in 14 years! Statista predicts that this trend will accelerate and that data generated will more than double to 394 zettabytes over the next four years. 

AI and other technologies are being driven by this explosion in data. Not the Nasdaq share price. So trust the trend that we are just getting started when it comes to the data we store and the problems we’re solving. It’s too easy to get lost in a North American or European bubble and assume globally that we have reached peak technology. We’re not even close. 

Secondly, be clear on how much time and money you are willing to risk to explore possibilities. This is a gut instinct, subjective call that will be made by the board and C-suite. There is no quantifiable mechanism for determining this objectively. The senior management team needs to decide on a decent budget, a fair timeline, and live by the consequences.

Companies often lack a good blend of domain and technical expertise. A recent development is that organisations try to find a jack-of-all-trades: traders who can code or coders who can execute trades. This strategy will work for some businesses but, for the vast majority, it won’t. 

Instead, it is important to look for experts in both fields who can work together. The best trader is highly unlikely to be the best coder and vice versa. They require different skills and mindsets. No one is hiring Ronaldo to play rugby.

The sheer lack of public knowledge about the commodities industry is emphasised by the popularity of books like “The World for Sale” or “King of Oil”. Though they are both entertaining, neither book is written by former traders. And so neither book can explore a trader’s approach in enough depth: how a trader forms a market view; chooses a position to reflect that market view; and then manages risk to take profit on that market view. Instead, these books focus on corruption and big bonuses. 

So given how secretive commodities trading can be, how do companies decide if they have the right technical and domain experts on board? The most important thing to hire for is curiosity. It’s key to leveraging cutting-edge technology in new ways. 

As an example, technical teams must be able to show curiosity for how a ship gets chartered or how cargo gets sold. Wanting to absorb this knowledge is really important because there are too many potholes for tech teams to fall into. They could spend weeks or even months on what they believe are valuable insights, but it will only take a domain expert a few seconds to recognise that they’ve taken a wrong turn. If a technical team is curious, it can save a lot of time and money. 

Equally, having a broad understanding of domain expertise is vital. Commodity trading and shipping markets are complex and full of nuances. Too many companies attempt to rely on individuals with very narrow expertise in one specific area of one field (e.g. derivative trading in one commodity in one geographical niche). These types of traders have almost zero curiosity outside of their niches (arguably, in their lives!) and they will not be able to envisage new ideas or approaches to build with the technical team. 

Don’t settle for mediocre team members because they will not have the spark required to leverage new technology. The types of people you hire can really influence your success in terms of using commodities data.

A massive error that many companies make is bringing in highly skilled technical people and then relegating them to the IT department, so to speak. If you look at the careers pages of some of the largest trading companies, you’ll see that they’re hiring data scientists, engineers, Python developers, and so on. That’s great, but you need domain experts – either internal or external –  to nurture them and act as a translator. 

At the end of the day, even the best coder, data scientist, or engineer, will never be able to add value to the business if they don’t have someone to share domain expertise with them. The most successful trading businesses have a cohesive approach where traders and technical people work together because when they operate separately, it simply doesn’t work.

Equally, traders need to be open-minded and flexible enough to listen to the tech teams in return. There is usually some resistance from the trading teams. Many traders might not see the immediate value in educating technical people about their market and sharing their domain expertise for two reasons. 

Firstly, when it comes to technology, it can sometimes be difficult to quantify ROI – nothing is guaranteed. Secondly, many domain experts rely on market assumptions built up over years of experience. When you start to use data and technology, these market assumptions are often challenged and can even be totally disproven. That can, and does, threaten egos. Not every domain expert has the agility to accept the limitations of their own mental models. 

And yet, data science and technology have consistently delivered impressive returns across all industries, from medicine to linguistics. So, with the right people and right approach, it’s very unlikely that within your commodities company, cutting-edge technology and data won’t yield a substantial return. 

Merge trading and tech teams and make sure that they are rewarded substantially for leveraging technology. The more that this tech is aimed at the revenue generation side of the business, the better. Technology is scalable in ways that many other business models are not and so it offers too good an opportunity for growth.

Before deploying technology, you have to be very, very clear about where that starts and stops within your business. A lot of companies make the mistake of trying something for a short period, and then quickly calling it a failure. Or they start to blame the technology for decisions that are separate to it. 

Any company consists of many different roles and employees. However, it is not always clear where the decision-making lies within the structure. There are plenty of occasions where, no matter how many processes have been created, things fall between the cracks. Businesses are an amalgamation of human biases, human decision making and human prediction. 

When it comes to data and technology, human influence can be all too prevalent. Which data to use? Which problems to solve? How to manage implementation? What to monetise, how and when? Cutting-edge technology is a bit like the sun. It’s powerful and you need to understand what it can and can’t do. But it will also shine a light on other systemic weaknesses within the organisation. So it’s crucial to make actual business decisions before you start deploying things built on data and tech decisions. As explored in a book called “Prediction Machines”, the more we rely on machines, the greater the value of human judgement. 

When it comes to trading, the most important business decision is for companies to define what they’re willing to risk. Otherwise, it’s a bit of a moving target where risk appetite constantly shifts – one week, everyone’s bullish, willing to risk everything, and the next, everyone’s bearish. Understanding how much risk you’re willing to take on to reach your goals, and the timeline of that risk appetite – be it a year, or even five – will inform your approach to technological adoption.

The data and tech landscape is changing too fast to standstill. Even if companies can find competitive advantage today, there is no guarantee that they can sustain this for tomorrow. Business models are constantly being challenged by outsiders and start-ups that can leverage technology in novel ways. 

To be clear, this goes way beyond tech tools. The technology itself is shifting how we interact with ourselves and others. As Botsman, the trust and technology expert, explains, we used to rely on technology to do something. Now, today, we trust AI to decide something. This is a dramatic shift in human dynamics. A huge “trust leap” as Botsman calls it. 

So how can a company stay ahead? Firstly, monitor what you have built. Leveraging data often means leveraging the relationships within that data. If the relationships change dramatically, the insights are no longer valid. For example, if jet fuel could suddenly be made from sand, then the price of oil becomes less relevant. A human expert will find out this information before a machine will. 

Secondly on a corporate level, understand that the traditional boundaries of businesses have disappeared. Companies are no longer just building something and selling it to customers. Instead, they can be suppliers, producers, end users, and collaborators, all at the same time, with multiple types of counterparties, in ever increasingly complex networks. So leverage that network in creative ways. 

Finally, be prepared to look crazy. Perhaps that dataset will yield nothing. Perhaps that idea will never work. Do it anyway. The potential upside is too great to ignore. And time is of the essence.

Want to learn more? Jack now runs a strategic advisory firm, Tradavex, where he helps companies grow revenues by combining commodity industry expertise with cutting-edge technology.

And for the whole “hire the right people” point, that’s where Cititec Talent can help!

Looking for more insights?

Get exclusive insights from industry leaders, stay up-to-date with the latest news, and explore the cutting-edge tech shaping the sector by subscribing to our newsletter, Commodities Tech Insider. Interested in being featured in a future spotlight? Reach out to [email protected].