EXCLUSIVE: ‘The Data Diggers’ – Mike Kiersey, Boomi in ‘The Fintech Magazine’
Boomi pioneered the concept of the integration platform as a service… but if you want to bring the data pieces together, you first have to find them, as Chief Technology Officer Mike Kiersey explains
How a bank goes about integrating technology systems in order to unify disparate data is pivotal to its digital transformation strategy, since it’s data that drives the modern bank.
That can create an enormous burden on IT, having to custom code and manually transfer files containing data that needs to be accessed by different process flows – for example, when creating an omnichannel experience for customers in which both they and staff can access the same information in real time across different channels, such as mobile and desktop. And when staff dealing with those customers need to access different data silos inside the bank. We’ve all experienced the all-too-familiar scenario of an apologetic customer services team playing pass the parcel with a customer because an agent can’t see all the information required to address their query. That’s where an integration platform can help.
Mike Kiersey is the chief technology officer for Boomi, which provides drag-and-drop integration and data mapping for a wide range of industries, from transportation to healthcare, education to retail and many more, over its Cloud-hosted, multi-tenanted AtomSphere platform. Legacy layers of technology and related data silos are clearly not unique to financial services; they are a fact of life in any environment that’s gone through progressive digital change.
When it comes to financial services, it’s often not the core banking systems that are the problem, says Kiersey.
“It’s actually some of the associated services that sit around them. Some of the bigger, established banking organisations have grown up with the Roman road approach: they’ve built on top, on top and on top of different technologies for point solutions, through multi-generations of technology change. And some of that needs to be dug out to allow a better transformation and ease of accessibility to some of those platforms today.”
As a pioneer in integration platforms-as-a-service (iPaaS) – recently also made available as a pay-as-you-use service, AtomSphere Go, on the AWS Marketplace – Boomi is focussed on what might be called data archaeology… helping organisations to find where these priceless nuggets of information lie – exposing them, but not necessarily moving them.
“Banks now have data in software-as-a-service-based solutions – they might have their customer relationship management (CRM) system hosted there; on-premise, such as reporting or business applications; and they have legacy. And there are a multitude of different ways that all those connect,” says Kiersey. “With the CRM, it’s through APIs to the provider of the service, but, in the middle and backend of it, the bank is still batch processing. So, it’s picking up files, putting it through a database and then into a mainframe. We’ve got to speed up the way data moves through the chain.”
However ‘smart’ a bank’s technology is, there will always be data silos, says Kiersey: “It’s the very nature of how applications have evolved, grown, been procured, because they’re solving particular business problems.” Rather than fretting about it, what banks should be focussed on is the metadata that’s attached to the information – the clues to its origin, content and location. “Using metadata is a better way to search, connect and then deliver new experiences, without moving the data itself,” says Kiersey. “It’s just referencing where the information should come from, which allows those data pools to be used effectively.”
That includes erasing it, if requested, under the General Data Protection Regulation (GDPR) – how, after all, can banks comply with a key tenet of the regulation, the right to be forgotten, if they can’t say with certainty where every trace of a customer’s data is? Many, he suggests, are paying lip service to privacy-by-design, because they don’t have a strong governance policy in place for the management of data lineage –the up-to-date filing of metadata, which avoids the misnaming, poor record keeping and duplication that leads to the most accurate files ‘getting lost among old or wrong ones’.
“If you take data out of the Cloud and put it on-premise to use, for instance, it’s still in the Cloud but now you’ve also got it on-premise. So, how do you keep them in sync? Having lots of data lying all over the place obviously incurs a huge amount of cost, management overheads and complexity. So, it’s not about putting your data in one place and then pushing it down to the legacy systems; it’s how you make it easier for those systems to be able to access a myriad of different data sources with ease and simplicity.”
In an increasing number of organisations, responsibility for a metadata programme ultimately lies with the chief digital officer (CDO), a role that was unheard of before a bank – America’s Capital One – created the first post in 2002. By 2012, still just 12 per cent of major organisations in the US had CDOs; six years later it was nearly 70 per cent, as data volumes grew exponentially and companies struggled to figure out how to handle them. Nevertheless, in a survey of CDOs last year by Informatica, 88 per cent admitted that many aspects of managing metadata were a big challenge, including data discovery, mapping business glossary terms to technical metadata, and mapping data lineage, process flow and proliferation. The survey also revealed that CDOs are seriously overworked, with multiple responsibilities for supporting operational efficiency, customer satisfaction, data privacy, innovation and revenue. An iPaaS solution would, then, seem an obvious choice.
Kiersey has a lot of sympathy for CDOs and he sees them as being critical to organisational change, standing as they do at the nexus between operations and IT and providing data-driven metrics to the board.
“It’s hugely important that both sides (operations and IT) have a greater understanding of metadata; that they know how to leverage it, how they can apply machine learning to make it more hypersensitive to both the employee – around what they’re doing in their job – and for us as end consumers. IT and rest of the business are not two separate functions,” he says. And when that’s acknowledged by an organisation, he’s seen ‘innovation starting to seep through quite aggressively, as it builds that common language and taxonomy around what the business is trying to achieve, and how IT can play a part in that’.
To help address the wider challenges of transformation, Boomi has recently linked up with Cloud-based core banking provider Thought Machine, looking at how itcan help customers move from monolithic mainframe to Cloud-native platforms.
“Those are the ones that are going to get the acceleration of speed and benefit from modern architectural practices and principles,” says Kiersey. “Then, they’ll become truly data driven, which means handling more data than they have already, yes, but they’ll be able to link it up with ease and speed so they can gain insights around user activity, what’s happening around their different product sets, and how they’re being consumed. This, longer-term investment in digital transformation will really aid the visibility of data.”
While the rise of Cloud-based, composable banking, plugged into an ecosystem of third-party providers, will trigger more data that demands better management, it could also generate an explosion in the use of APIs, the mechanisms through which the data is exchanged.
“One financial institution I’ve talked to recently has about 2,000 APIs,” says Kiersey. While that’s impressive, it raises similar questions around management: “Do people understand what those APIs are, where they are, what they can do with them and therefore what they really mean to their business?” he says. “Because we’re only ever going to increase complexity by using APIs, unless we have a true strategy for managing them, understanding what they mean to the business and how we’re planning to link them up across the organisation.” API complexity is a consequence of a microservices environment that Boomi is already helping some of its 15,000 customers to address through its suite of integration solutions.
As the so-called late-majority of organisations, who delay adoption of innovative technology until the pioneers have deployed them successfully, get on board, Kiersey has this advice for banks: “We’ve all heard of the tipping point, where small changes ripple up; once you’re over 51 per cent, you’re into big change, then epic change. Banks need to look for those short-term wins that reduce cost, lower risk, and improve simplicity and productivity. If they take that small, baby step approach, once the value starts to be seen, confidence will manifest for the longer term. If they can do that in the right areas, where cost has traditionally been seen, they’ll start removing some of that technology debt and, over time, they’ll start to simplify the underpinning architecture.”