Old school monitoring was designed for techies. Often focused on assessing server and network health, it’s still extensively used today to enable infrastructure groups to detect and subsequently fix developing issues that could impact trading, such as a queue that’s building, a full disc or network delays. All of which is very valuable if you’re responsible for resolving problems from a technical perspective.
However, the sort of information often provided offers limited insight, if, for instance, an incident has taken place and you’re trying to quantify your business’s exposure, understand if you’ve breached a regulatory commitment or proactively manage the experience of impacted clients. To address these types of requirements, you really need to know exactly which orders and trades have been affected, how and the clients they belong to.
Technological progress has now led to the development of more contemporary monitoring approaches that focus less on infrastructure health and more on assessing the real-time status of the actual orders and trades moving across business processes, such as trade execution or exchange matching procedures. So users can detect changes to the ways in which these business items are flowing that could impact performance, the business or the end-user’s experience.
These approaches still provide technical teams with the information needed to identify and resolve the root cause, but additionally they offer the actionable insight necessary to manage the issue at a business and customer level. In doing so, delivering the potential for real trade profitability and client experience benefits - ultimately expanding the technology’s relevance very firmly into the business sphere.
Innovators blaze the trail
It’s the industry’s more innovative players that have been at the forefront of adopting this new style of monitoring, coined business flow tracking. However, over the past 12 months mainstream interest has started to emerge from firms facing:
- Growing regulatory pressures in the form of initiatives such as MiFID II and the Senior Managers Regime. Both of which take the concept of responsibility, and where it lies, to a whole new level - driving the requirement for greater operational oversight
- The need to more effectively minimise trading losses caused by poor quality market data or a trading system’s unexpected response to market volatility
- And increased client demand for improved service levels and greater visibility regarding what’s happening to their orders and trades
Moving up a notch
These firms are looking for ways in which they can reduce the amount of time they spend firefighting and are keen to introduce a means of more reliably predicting, proactively managing and controlling issues. So they can start to improve service levels and derive greater business and client value from their monitoring solutions.
For many, their search starts by understanding where their current approach really lies of the business value spectrum and determining where they want to be:
Infrastructure health monitoring
Checking if servers, networks and processes are up and running is a standard, and often very valuable, activity performed by many firms. However, these components are often assessed in silos so if a firm only employs infrastructure health monitoring, understanding what’s happening across complete processes can prove challenging. Also users often don’t understand the business context of the data being transported. So when an incident occurs, whilst technical teams can detect the faulty component, determining the issue’s knock-on effect on the orders and trades being processed, is very difficult.
So the next stage is to be able to examine the actual data moving through the process. By capturing, for instance, a market data tick at multiple points and recording the hop-by-hop timings, when an issue occurs users can investigate if delays are taking place at a certain point. This type of monitoring facilitates basic issue alerting and enables users to react to problems in simple business scenarios. And there lies the limiting factor – end-to-end trading processes are rarely simple.
To really grasp what’s happening as an order moves through a business process involves fully understanding the data linage (the various changes and hops it goes through from origination to completion). So tracking and analysing a trade involves determining which market data tick generated the algo trading decision, the market orders created and all resultant fulfillments. This can help answer business and regulatory questions such as:
- Why did this trade lose money?
- Did we provide best execution?
- Or why did the algo engine decide to execute this trade?
By chaining the related data together every stage of the entire trade can be examined and compared to past trends. Therefore, if current trades are proving slower than previous trends suggest they should be, engineers could be notified that degradation may be starting to emerge and gain a head start on proactively investigating the situation.
With this approach, we’re moving into the realm of business flow tracking, providing a comprehension of how business processes are working in reality, the ability to detect when they are not and accurately forecast the potential consequences of process changes.
Assessing data quality to improve service levels
Next firms look at the quality of the data traversing these processes – its timeliness, correctness and completeness. Being alerted to a divergence from the norm can provide the actionable insight necessary to curb financial losses. For example, by detecting quality issues starting to impact the data used to formulate algorithmic trading decisions, firms can stem the data’s flow to avoid the rapid accumulation of expensive trading losses.
Furthermore, by measuring the data’s quality against trends over time, client SLAs or regulatory commitments, firms can continuously assess whether they are meeting these obligations.
Using monitoring to drive business and client value
Firms operating at this level bring together all of the information collected by the previous stages. Doing so, enabling them to understand what’s happening across business processes, measure performance and service quality against client commitments and use this insight to drive real business and client improvements.
It’s no longer about just measuring whether SLAs are being met but using predictive analytics and actionable insights to increase the likelihood of achieving them.
Taking a cohesive and intuitive approach firms can compare current performance to past trends, foresee upcoming client SLA and regulatory breaches, and take the action necessary to avoid an issue occurring. And when an incident proves inevitable, access the insight necessary to proactively notify clients, and keep them updated to effectively manage their experience.
Still think monitoring’s for techies?
Whilst previous approaches to monitoring addressed the needs of technical teams, recent advancements definitely enable firms to achieve more business-relevant value.
Monitoring is no longer a binary question. It’s not whether you have it or you don’t, it’s more a case of assessing whether your current approach is delivering what’s needed to meet regulatory, profitability and client demands. And if it isn’t, exploring how recent advancements could help get you where you want it to be.
If you would like to learn more about the potential of new monitoring approaches, please download our white paper 'Beyond Monitoring: Where is your solution on the business value spectrum?'.
By Steve Colwill, CEO, Velocimetrics.