By Andrea Novara, Engineering Lead | Banking & Payments Business Unit Leader at Agile Lab
For years it has been a given that financial institutions must manage and retain large volumes of digital data securely for compliance and legal purposes. Whether for specific industry regulations like DORA, or for privacy obligations such as GDPR, these legal requirements must form part of an overall governance structure. However, as the regulatory load has become heavier, organisations are finding it more onerous and costly to protect their data while also releasing its untapped potential.
Trapped within these vast – and often disparate – data lakes, warehouses, archives, and storage systems, lies considerable value. Many contain detailed information about customers, buying patterns, products, and market performance. If these insights could be accessed easily and utilised effectively, they could empower substantial business growth. From driving new product innovation and enabling faster reaction to market changes, to creating highly personalised, customer-centric offerings, or even informing strategic planning, the scope could be huge.
The burden of data management
Instead, data management is becoming a financial burden, holding back creativity rather than stimulating new ideas. This issue is only set to get more challenging with the proliferation of AI-powered applications creating unprecedented amounts of unstructured data.
Attempts to unshackle data in the past have met with limited success, restricted by the legacy infrastructures of long-established banks and financial institutions. Having expanded through multiple acquisitions, many have disparate data repositories reliant on incompatible technologies and managed by different functions and service providers.
Expensive vendor lock-in often prevents organisations assimilating data into a common format. Not only are the high costs of consolidation unpalatable, but the disruption to business would also be equally unwelcome. A wholesale change of technology is simply not feasible or desirable for most and, until recently, was the only course available.
Computational governance explained
This is no longer the case with the advent of computational governance, which is bringing much-needed automation to the enforcement of data standards and compliance across enterprises.
As an overarching approach, automated governance can span all business units, functions, and geographies, bringing consistency and reliability to the data management process. It encompasses all areas where standards must be followed, including data quality, integrity, architecture, compliance, and security. Above all, it is technology-agnostic. Deployment does not require upgrades to existing computing and storage infrastructures or data formats.
Importantly, it should not be confused with traditional data tools that create new data or consolidate and deduplicate existing repositories for further processing. A computational governance platform doesn’t generate data. Instead, it provides a vital governance layer, translating external regulations and internal policies into automated guardrails when data projects are initiated. Rather than reacting to compliance errors further down the line, this prevents mistakes going into production in the first place.
The governance system cannot be by-passed, ensuring all new projects are always compliant. Existing products and services can also go through the process retrospectively to understand what’s required to bring them into line with current standards.
Defining data as-a-product
By enforcing guidelines consistently, product planning and decision-making can be made on a reliable footing. Automated, customisable templates help product and data owners build projects quickly, and as a result, the effort needed to compile information is drastically reduced, saving up to 50% of time previously spent on finding and verifying data and creating more coherent results thanks to the best practices embedded into the templates.
For product managers in the finance industry this could be transformative. No longer will they need to build new products and services from scratch, as the approach enables users to extract data from existing sources without additional effort or incurring unnecessary costs. Data can be redefined into products that can be utilised by other departments, teams, or geographies.
For example, where a bank is made up of multiple acquisitions with differing legacy technologies across countries, the system will enable customisation of financial products to specific local rules, while ensuring they are all described and perform in the same way. Thus, a product can be rolled out to different countries but with a single, common interface. This could be anything from consumer and business-to-business offerings, such as credit lines or global payments, to internal financial planning and reporting requirements.
Unfettering business agility
Computational governance represents a new way to ensure that data adhere to quality standards. It enables organisations to move away from siloed repositories and technical teams operating independently. This user-friendly approach to data consumption enables domain experts to unlock and share the potential of their data assets. Moreover, it puts the onus on the automated governance system to uphold external regulations, such as GDPR and DORA, and internal standards set by the compliance team. This takes the responsibility off the shoulders of developers and data users.
This in turn allows wider access to data, enabling greater agility and responsiveness, while reassuring business leaders that strict compliance standards are being maintained. Notably, it can be applied to new technology solutions and AI-powered tools, ensuring investments made now are futureproof.
The idea that compliance can enable agility may seem surprising, but computational governance is proving the point. Capable of slashing time to market for products and services, it is a game-changer for financial institutions previously weighed down by complex data issues and regulatory requirements.
