Data science will be the linchpin of future personalised banking services, Bank of Ireland’s Colin Kane tells John Kennedy.
The kind of personalisation we all take for granted when a platform such as Netflix surfaces shows and movies we might like is informing the future shape of banking services, too, according to Bank of Ireland.
Banks are turning to data science, machine learning and artificial intelligence to re-establish that personal touch and intimacy that has been challenged by the era of self-service and open banking. The result will be relevant, timely services and less friction through the use of real-time data, predictive engines and blockchain.
‘The customer journey is becoming more complex; there are more devices and channels’
– COLIN KANE
Colin Kane, who is responsible for analytics and data science at Bank of Ireland, explained that as the customer journey becomes more complex due to the myriad of devices and channels, re-establishing that intrinsic relationship and context is crucial.
The future of banking, he said, will be defined by financial players who can accurately meet the needs of customers depending on what’s happening in their lives.
Kane explained that data science is already a core fixture at Bank of Ireland, where various teams of data scientists and data experts are involved in analytics functions such as risk analytics, anti-money laundering (AML) fraud and the area he himself works in, marketing and customer experience.
“My area of expertise is in marketing, customer analytics and customer engagement, using data science and analytics to understand how customers behave and interact with customers.”
He explained that the banking world is moving away from traditional marketing strategies to more targeted and effective tactics. “All organisations spend a lot of money on marketing and it is always the same old problem of trying to understand which activities work and which don’t.”
He said that the old standard of a one-size-fits-all approach to marketing across traditional media is falling by the wayside in favour of accurate marketing via mediums that reach the right target audience at the right time. “The customer journey is becoming more complex; there are more devices and channels. How do we work out which ones are impacting our sales?”
This requires Bank of Ireland understanding complex data models and datasets, mapping out those journeys, and running simulations to understand how each channel affects specific tribes of customers.
“Where are those channels influencing the digital journey or the sale? Is it a channel we need to increase our spending in, and what can we expect in terms of delivering additional sales? A lot of it is around bringing credibility to marketing. The business of advertising and promoting products is less and less about art and more and more about science. This affects everything down to the length of subject headers in email.
“The core of what we do is, we study customer behaviour and how they interact and respond. A good example would be how you prepare ads for car loans. A person looking for a family car will respond to a message that emphasises safety. A younger person might be more influenced by style. The next bit is, how do you turn that into a real-time interaction? Can you automate engagement?”
Kane said that in the past, there were just a few channels that marketing staff at Bank of Ireland were concerned about, including print, radio and TV. Now, it embodies more than 15 channels, including social media, which all need to demonstrate interaction and engagement for effectiveness.
“How do we understand where we are getting our bang per buck? That’s the key focus we have today.”
Kane said that the future of banking will be driven fundamentally by using data science to predict consumer behaviour with greater accuracy. “We have to work within the principles of the EU’s General Data Protection Regulations (GDPR). While that may seem like a challenge, there is an opportunity to use customer data in a way that is similar to organisations like Netflix, to advise customers with the right action at the very moment they need it.
“This involves creating decision engines and promoting the right actions at the right time, such as when is the best time to offer a customer a credit card? A customer could be advised if their savings would deliver better interest in a different kind of account.
“Crucially, however, the motive is to not always sell products but also identify an issue and help the customer and provide banking that is much more in tune with their needs.”
Bringing banking into the cloud-first world
As open banking grows and new fintech players enter the fray, Kane said that the challenge for Bank of Ireland on the data side is bringing together the right data across all the different touchpoints – from the online banking website, the mobile app, the ATM, the contact centre, and wherever and however a customer chooses to interact with their bank.
“We have good data warehouses in place, strong demographic data and, because the customer is moving faster between different devices and channels, good behavioural data. The challenge is storing it and accessing it in such a way to develop and build models that ensure we provide proactive and relevant services that meet their needs at the right time.”
Another challenge, Kane said, is Bank of Ireland choosing the right cloud technology to be cloud-ready and cloud-first. “Banks are heavily regulated and we have to be careful about how we manage customer data, and this also means using Amazon Web Services and Google Cloud in a much more efficient way.
“When dealing with an organisation of our size, we have to be careful about the transition to the cloud-first world.”
Blockchain technology presents an opportunity for banks to create indelible footprints of transactions but also to bring legacy applications and systems into the cloud environment. “In terms of blockchain, we are looking at a couple of different use cases, particularly around how customers engage with different processes, documentation or moving through different steps of a process.
“One future application could include mortgage applications and handling that frustrating stage where there is a lot of paperwork, and ensure a smoother process of mortgage onboarding,” Kane concluded.
“At the moment, these are proofs of concept rather than use cases but the bank is definitely looking for opportunities in blockchain as a way to manage this and take frictions out of the banking experience.”