The cloud has the potential to evolve workflows, impact product roadmaps and support planning and budgeting, but some FIs are uncertain about which applications to run in the cloud and which to maintain on premises.

Finextra spoke to Tom Dejonghe, senior director of SaaS transformation at Collibra, about its recent report ‘Power Digital Transformation with Cloud Migration’ and the benefits of the cloud as industry confidence in the security of the cloud grows.

Disruptive decision cycles

Dejonghe highlights that historically, FIs enjoyed the luxury of longer decision cycles. “You had the ability to make mistakes and drive big decisions, but your ability to navigate that was a little bit more forgiving because of the time you had to do that.”

He explains that recent disruption has resulted in the taxi vs. Uber debate or Blockbuster vs. Netflix, for that matter, and data has driven this transformation. “We’re seeing this type of disruption across all industries that were traditionally slow-moving and conservative.

“Data has helped companies like Uber and Netflix disrupt and become successful companies because if you’re not extracting the intelligence you need from your data, at almost near real-time, someone else will do it and they will come after your business,” Dejonghe says.

Navigating compliance and caution

Technology vendors have long navigated the conservative culture in financial services that has been instilled by varying levels of compliance, which has in turn, resulted in a cautious approach to integrating and adopting new technology.

Due to the financial crisis, even more regulations around data were introduced (BCBS239) and more recently, around data privacy (GDPR and CCPA) in an already heavily regulated industry of financial services. This makes it particularly hard for technology vendors to navigate an inherently risk-averse culture around adopting new technologies. However this is changing due to competitive pressure from new players on the horizon such as PayPal and Facebook (with the Libra), and obviously because cloud is considered to be a primary choice for many enterprise businesses.

Dejonghe adds that in addition to this external pressure, there is increased awareness about cloud security with many FIs beginning to realise that data can actually be safer in the cloud than in their own data centres, since cloud providers invest in security practices and controls at a scale that most organisations cannot. For instance, AWS implements global security and compliance controls, meets high standards for privacy and data security and offers the largest network of security partners and solutions. AWS data centres are secure by design, and its global infrastructure is built for resiliency.

Dejonghe continues: “Security and compliance are a big part of legacy culture, as well as conservative decision-making and organisational culture. In enterprise-sized organisations, hierarchy still plays a big role in the decision-making processes. Newer, data-driven players have leaner hierarchies and therefore, greater agility to make faster decisions and have shorter feedback loops. FIs are also adopting this lean and agile culture.”

Decentralising the development lab

The Collibra report states that “migrating to the cloud does not happen in a vacuum” but is part of a broader digital transformation initiative that relies on DevOps and other development strategies. These are fast-paced environments characterised by cross-functional teams that execute overlapping and never-ending cycles of development, testing, remediation and improvements.

“Data engineers and data scientists can quickly concentrate computing resources and spin up high performing environments in the cloud,” Dejonghe explains. “Cloud can support massive data lakes and empower developers and other personnel to work in unison, whether they are in the same building or spread around the globe. It’s like a decentralised development lab on demand.”

On the subject, he points out that not everything should be centralised. “To optimise for performance but also for security and compliance, a distributed network of edge compute, which lives in the proximity of the data, can save a lot of time and reduce the need for error-prone data transformations and extractions. We adopted the centralised model around the cloud at first and now we are rolling back to more of a hybrid distributed model.”

A single view

The Collibra cloud platform, which runs on AWS, provides what Dejonghe refers to as a “single pane of glass view into the data landscape,” but this information must be aggregated to ensure that the intelligence around data can be created. “That single pane of glass allows data engineers to use APIs that can help build capabilities into solutions, complementing the products that are already in market.”

He continues: “In the world we’re living in, despite knowing where the data is located, it is difficult to ascertain who owns the data. If this is unknown, you do not have the ability to trust the data and cannot validate the information. Data should be trusted.

“That single pane of glass allows you to control your data, see who owns the data and understand the accessibility that the cloud provides. There is no need to go through a long workflow and request permissions. The notion of self-service and disintermediation is really accelerating and is enabled by the cloud and compute models, governed by and made accessible through a data catalogue Dejonghe concludes.