FX data analytics: banks’ new frontier
Banks cannot afford to ignore the potential value of using insights from data analytics to personalize their FX services.
For example, tailored pricing based on a variety of data analytics can result in greater trade flows and enhanced profits.
Based on historical behaviour and related to different criteria, such as product or currency, these analytics can help provide answers about a client’s previous trading history and whether it is following an upward or downward trend.
Last year, Deutsche Bank announced it was partnering with FX exposure management solutions provider FiREapps to enable the bank to offer data analytics to its corporate clients.
The solution offered analysis of corporates operational data, working within the enterprise data systems and processes to aggregate, validate and analyse underlying exposure data.
Deutsche declined to comment on how this service impacts its FX offering.
New products require discussion by committees, rigorous compliance processes – it becomes a multi-headed beast
Nigel Farmer, solutions director, capital markets at Software AG, explains that analytics can identify the conversion rate to executed trades and establish if it is increasing or decreasing, providing information on whether the client trades when the market moves and how close to the bid/offer banks need to be to turn a request for quote into a trade.
“The same analytics can also demonstrate the profit/loss profile on the trading or hedging of clients’ trades,” adds Farmer. “However, the high volume and frequency of trading on the FX markets means that these analytics need to be accurate, current and streamed in real time.”
According to Duncan Ash, senior director, global financial services at Qlik, all banks offer bespoke FX products and services to their customers to some extent.
“The more sophisticated are using analytical insights derived from data to make sure that those products are priced attractively to customers, but are also well-balanced from a risk perspective,” he says.
“It is easy to offer competitive pricing, although it requires much more skill to do this in a way that doesn’t present a greater risk profile to the bank. Risk-adjusted pricing is one of the major benefits of using a data-driven approach to pricing.”
Using traditional analytics and business intelligence tools, Ash says it might have taken between 12 and 18 months to implement the necessary systems, which proved a deterrent for some banks. The resources required in terms of both software and manpower, and the necessity for business processes to change to work around the software, were further reasons for reticence.
He suggests the cost of implementation is no longer prohibitive.
“By adopting a modern approach to visual analytics, a customer could have a prototype running in a couple of weeks and be in production in less than months,” says Ash. “It is not disruptive and can be delivered with a much smaller team.”
Banks looking to personalize services in FX need to be able to identify behavioural markers such as strong purchasing patterns, which will allow them to offer denominated accounts or currency offers, explains Vincent Kilcoyne, capital markets lead, SAS UK & Ireland.
“However, one of the biggest obstacles to a personalized product approach is not knowledge of the customer – it is the agility required to create new products,” he says.
“New products require discussion by committees, rigorous compliance processes – it becomes a multi-headed beast. More realistic for most banks is the ability to offer existing products, in response to the behaviours they observe in customers.”
Kilcoyne accepts that implementation times for personalization projects vary depending on how data-ready the bank is. The starting point is a single view of the customer, and data quality is the biggest impediment to this, so banks need to ensure customer data is of a good enough quality to begin drawing insights.
He observes that implementing real-time analytics for personalization can have a number of quick wins and multiple implementations can be put in place to be scaled up at a later date.
“Integrating data from phone calls and emails can have huge benefits in terms of understanding customer preferences,” says Kilcoyne. “Implementation can be rapid, but adjustment can take more time. Introducing newer, more disruptive thought processes requires a change of mindset.”
Trying to re-engineer old storage, mining and management technologies to achieve this kind of insight can be hugely expensive, adds Kilcoyne.
“New technologies make rapid processing, storage and incorporation of unstructured data much faster and cost attractive, although some organizations view the adoption of these new technologies as a huge learning curve,” he concludes.