Trading as a Service in intraday power markets
An interview with Jürgen Mayerhofer, CEO and Co-founder of enspired
ComTech Advisory: enspired is a relatively new player on the market. Can you give us a quick overview of the opportunity that you see and the services you are offering?
Jürgen: We are in the middle of a fundamental shift in short-term energy trading. As you know, I focused on algorithmic trading and intraday power markets over the last years. There is one big topic that needs to be solved to make the energy transition happen – integrating flexibility smartly into the grid. This applies to existing assets and new projects alike, such as power generation, battery storage or distributed portfolios. With recent progress in intraday markets across all of Europe, electronic trading interfaces at the power exchanges and intelligent algorithms at our fingertips, we now have everything in place to make this happen.
enspired is all about unleashing that commercial potential of flexibility. We do this by providing trading-as-a-service for intraday power markets to our clients. Our services cover the entire spectrum from market access and fully-automated trading based on self-learning models to cutting-edge execution technology. Our mindset is that of a data-driven tech company that trades energy.
ComTech Advisory: Why is existing flexibility not monetised today, given that it represents an additional revenue stream for incumbent players? Shouldn’t they be eager to get into that market?
Jürgen: That question made us wonder for some time, and the answer depends on who you talk to. In my view, the primary three obstacles are:
Transparency: most companies don’t know the value of their flexibility, or they don’t believe the figures – this is something that I have also witnessed at the beginning of the algorithmic intraday trading trend. At the moment, trading services on European intraday markets are non-transparent and dominated by a few large players. We are going to change that.
Skills and resources: many companies don’t have analytics teams, the required technical infrastructure or lack software engineering capabilities. If you want to get the most value from your flexibility, you need to automate the whole process from data intake to trading execution.
Strategic focus: asset operators and small to medium-size players often have a different business model and therefore don’t have a focus on 24×7 trading. Some don’t even have direct market access at all. They rely on third parties for those activities, which impacts how much value they are able to capture from their portfolio.
ComTech Advisory: You already mentioned self-learning models and automation, will AI also play a role in intraday trading?
Jürgen: Definitely, yes. We see automation as a fundamental requirement, and this trend has been going on for quite some time now. Expectations around what you can do with AI are very high. A point in advance, it is not feasible to simply throw a pile of data into a model and magically get sound trading signals. You need to design the learning process carefully, choose the relevant data, evaluate and backtest signals and predictions. This data-driven approach needs to be at the heart of your trading operations if you want it to be successful in the future. As we are integrating deep reinforcement learning into our trading services, we can already tell that AI is changing the way power is traded fundamentally.
ComTech Advisory: It seems you try to coin a new term – augmented energy trading. It sounds quite similar to algorithmic trading?
Jürgen: I get this question quite often, and maybe it’s just because of my background in algorithmic trading. From what we see in the energy market, algorithmic trading solutions are mostly focused on automating the execution process, using rule-based order routing. Augmented energy trading, in contrast, is a much wider concept where algorithmic execution is only one piece of the puzzle. At the heart of it, it’s about augmenting the capabilities of our human trading experts with the use of state-of-the-art technologies, applying it to energy markets and thereby improving the commercial results of our trading services.
ComTech Advisory: Sound interesting, but how does it work in practice?
Jürgen: Let’s look at an example to make it more tangible: when optimizing short-term flexibility of a power generation asset, a common algorithmic trading setup is the following: you take the current market prices, which in reality oftentimes means historic data from a few minutes ago, feed it into an optimization engine to calculate a dispatch schedule, and then pass over the resulting volumes to an execution algorithm which places orders on the market. That process is sequential, and the data is usually outdated by the time the models finish calculating.
We turn this onto its head and integrate all parts into a single platform with data flowing back and forth:
- The intraday market is extremely dynamic, and changes in the order book can happen at any point in time. To achieve the best outcome, we evaluate how our own orders interact with the rest of the market. Executing a trade is a constant trade-off between price and execution probability, so our models feature parameters to dynamically balance between these two and use that as inputs for the optimization.
- The technical aspects of the underlying asset play a crucial role – producing more power now may prevent you from producing less at a later point in time because the power station has physical constraints. These are built into our trading systems as digital models, so our platform can learn how the asset behaves and decide intuitively, rather than having to calculate scenarios.
- We are forming expectations of future price movements and volatilities instead of looking at past information. This in itself requires a vast amount of data to be evaluated, ideally in real-time, for example weather updates, power consumption levels or system balances. Traditionally, trading teams have coped by adding more and more screens to their trading desks. We use software interfaces to feed this data into predictive models and perform feature engineering, which helps us select the most relevant information.
When joined together intelligently, you have a trading setup that provides value- enhancing functionality for short-term energy optimization that surpasses previous approaches. And our clients get all of this as a one-stop service.
About Jurgen Mayerhofer, CEO and Co-founder of enspired
Jürgen is CEO and Co-founder of enspired, a technology-driven energy trading service provider with a focus on European intraday markets. Jürgen has spent more than 17 years in the energy industry progressing from software engineer to senior leadership roles over the years. Most recently he was managing director of the algorithmic trading software vendor VisoTech, which was acquired by TMX/Trayport in 2019.
Keep in touch and sign up to our Newsletter
