Public cloud computing is early on its journey to core of the bank

Banks are no longer reluctant to use the cloud as the technology pierces the outer layer of the banking sector and begins its journey to the core

This article can also be found in the Premium Editorial Download: Computer Weekly: Microsoft vs Google – who’s winning the cloud email race?

Banks have quickly moved from not talking about what they are doing in the cloud to publicly boasting about the size of their clouds.

Traditionally, banks don’t like to talk about what they are doing with technology until it is established and proven. They don’t want to give away their secrets or risk damaging headlines if things go wrong. But at the recent SIBOS and Amazon Web Services events in London, they were not holding much back when it comes to cloud computing.

Why the change? Competitive tension and digitally savvy customers are two forces driving banks into the cloud in plain view.

In the past, apart from IT professionals and tech geeks, customers didn’t care that IBM mainframes were purring away in the banks where their money lived – and why should they? As long as their salary went in each month, ATMs gave them money and their cheques didn’t bounce, they were happy.

But today’s consumers want to bank in real time. They want to be able to watch what their money is doing, move it around quickly, pay for stuff in an instant – and all of this on their smartphones.

As a result, customers want to know about the technology that banks are using and their plans. It has become a competitive differentiator to be a tech leader in the face of increasing competition from digital-first companies disrupting the sector.

The emergence of these tech-led challengers, many of which exist only in the cloud, means consumers have more choice and have become more demanding. Millennials wouldn’t open a bank account with a company that doesn’t offer a full digital experience. It is this pressure from the customer base for products and services delivered through fintech that has forced banks to move into the cloud – and talk about it.

The cloud enabled these challengers to transition rapidly from idea to regulated finance provider. They can create products and services quickly through cloud development – but the cloud also allows traditional organisations to respond quickly to customer demand for products and functionality. And that is what we are seeing from big banks today.

Speaking at Amazon Web Services’ (AWS) financial services conference in London last month, the company’s financial services chief, Scott Mullins, used an Australian anecdote to bring home how far the acceptance of cloud computing has come in a short space of time – in this case, how banks use the cloud to develop products.

Mullins said National Australia Bank (NAB) has internal tools to help train 3,000 staff on how to use modern technology to create applications, and it is using the cloud to support this. “At NAB, staff no longer need permission to build something in the cloud,” he said. “If you want to build something on AWS at NAB, you can do it immediately without asking.”

A year ago this was not the case and permission was needed, he said.

Read more about financial services in the cloud

Examples of traditional financial services businesses using the public cloud to develop were also featured at the AWS financial services event – Goldman Sachs and Direct Line, traditional businesses in investment banking and insurance, respectively, have launched completely new businesses in the AWS cloud.

Last year, Goldman Sachs launched a business in the UK, known as Marcus, which offers savings accounts to consumers, whereas the bank is traditionally focused on institutional investors.

Joanne Hannaford, partner in Goldman Sachs’ engineering division, heads up technology in Europe at the bank that she joined 23 years ago as a junior programmer.

Software engineering is a huge part of the investment bank’s operations – with 11,000 developers worldwide, making up a large chunk of its 35,000 workforce.

Speaking at the AWS event, Hannaford said although Goldman Sachs, like its competitors, has been using cloud for a long time in terms of virtualisation and building private clouds, today it is building new businesses completely in the cloud.

Goldman Sachs’ Marcus platform was implemented in the UK, end to end, in 11 months. Hannaford said this would have been unthinkable in the past if the company had had to provision its own hardware, which alone would have taken five months.

She said that within an hour of its launch, Marcus had 600 times more customers applying than had been expected. “I don’t think we could have done that without a relationship with a cloud service provider like AWS,” she said.

Darwin in the cloud

UK insurer Direct Line is another example of a traditional finance business that has set up a new business completely in the AWS cloud.

The new business, known as Darwin, was set up earlier this year to provide a digital platform that could be quickly adapted to changes in demand.

Direct Line wanted to build a business that could change constantly to keep pace with customer demand. Still in its early stages, Darwin currently only works through price comparison website Moneysupermarket, but will expand to other price comparison sites.

Darwin is built in the AWS cloud and only a handful of people are employed in the unit. Sumit Bahukhandi, director of corporate ventures at Direct Line and founder of Darwin, said it uses the Amazon “two pizzas concept”, where if a group cannot be fed by two pizzas, there are too many and nothing will get done.

For anyone attending a conference held by cloud giant AWS and Swift’s annual SIBOS get-together in consecutive weeks, it becomes hard to imagine any financial services firm not already deeply engaged with the public cloud.

But there are many levels of cloud adoption. During a cloud discussion at SIBOS in London last month, a snap poll of an audience comprising hundreds of bank executives revealed that 44% are already in the public cloud, 35% are catching up, 19% are considering it for the future and only 2% do not have it on their radar.

The snap survey also revealed that 37% are using the public cloud to complement on-premise systems with artificial intelligence and machine learning capabilities, 35% as alternative to non-critical on-premise applications, and 28% for on-premise mission-critical applications.

And the spread of the cloud in banking is not determined by the level of IT resources that large banks dedicate to it. Beyond what the likes of Goldman Sachs and tier one banks are doing internally, there is an ecosystem of services available to banks via the public cloud.

Anti-money laundering

A good example is anti-money laundering technology and services. Banks worldwide face huge fines if they fail to put the right processes in place to ensure they are doing their upmost to trace, and ultimately block, money laundering in their customer base.

This is a huge problem, with only about 3% of money laundering activities successfully blocked. Banks have to address this better because regulators are hitting them with fines running into the hundreds of millions.

Banks cannot afford not to address this problem, but because current systems aren’t working, they need to try something different. Big banks in the Netherlands have decided to work together to set up a standalone operation that will combine huge volumes of transaction data with their vast IT expertise to focus on anti-money laundering, but this might not be possible for all banks.

For some, the cloud is the answer. The supplier community via the cloud is now an option. For example, tech startup Comply Advantage provides on-demand cloud services to banks, via AWS, including transaction monitoring. It has 450 financial services customers across 45 countries.

But all this is still only scratching the surface in terms of the IT engine that runs banks. You might say that the lights and windows of the car are cloud-powered, while the engine is still running on mainframes.

Mainframes still account for most of a bank’s processing. These systems are secure, cheap, reliable, and are understood by regulators. In fact, an IBM-commissioned study by McKinsey showed that although the simplest workloads are in the process of migration to the cloud, 80% of total workloads are still on-premise.

So if customers have funky apps that enable them to transfer money between accounts and pay for stuff using their phone, do they really care which systems process the transactions behind the scenes? Perhaps it is a lack of customer pressure that has meant the cloud has been slow to replace transaction processing on mainframes.

Eventually the cost of maintaining mainframes themselves, as well as the need for processing engines that can support the banks of the future, will change this.

Read more on Financial applications

Data Center
Data Management