The Importance of Data Quality in Open Banking & API Adoption

By James Briers, CTO of Intelligent Delivery Solutions

What is Open Banking?

Technology and banking have always had a complicated marriage. New technologies make processes easier, more efficient, reduce errors, improve communication, and change how consumers see and interact with money. It makes for excellent service, and it’s lucrative.

In terms of product development, the nature of customer spending habits can be analyzed to make better decisions, and organizations can base strategic decisions on real-time API metrics and data streams to respond to changing market trends.

With the adoption of more SaaS-based applications and digitalized services by more financial institutions, the opportunity for risk increases too.

The importance of data quality and security will only become more prevalent in a world where targeted attacks become more common and where failures in compliance are often spectacular and very public for banks.

API technology undoubtedly offers great opportunities for innovation in the financial services sector, allowing forward-thinking organizations to steal a competitive advantage.

Modern businesses are built on data, Banks are built on their customer’s trust. The instant that the security of a customer’s money – or their information – is put at risk, that trust is lost.

It’s therefore imperative to have 100% certainty of the data used in the development and implementation of any API tool that requires that data to function.

Implications of Open Banking API Benefits on Data Quality

Ease of Use Characteristics

Open banking trends have enabled customers to have greater control over their personal finances. They are able move and manage their money with assured security from strict regulations.

Open banking mediums are easily accessed, most predominantly from personal computers and smart phones, making them popular for end-users with limited technical expertise.

The customer experience has greatly improved through open banking. It has provided many benefits to end-users like cheaper credit cards, smarter banking methods, less intrusive checks, and more affordable insurance products.

Banks have the advantage of offering tailored products and customer service based on augmented intelligence algorithms predicting customer needs based on their behaviour.

Banks will subsequently have different methods of collecting and storing this data based on the core banking system setup and products offered to customers.

Therefore, any transactional data quality depends on how this data is initially structured. With so many different, complex information types being collected, this can become problematic for data cleansing processes.

Data teams will spend more time using potentially ‘dirty’ and disorganized data for analysis. Naturally, valuable information may be lost or not used, demanding more efficient data preparation to be executed.

Real-time Decision-Making

The immediacy of message-based transactions, delivered from APIs, benefits users at a corporate, partner and customer level.

Interfaces are built to deliver quick, standardized messages based on a predefined, technical definition encoded into the API. This standardized structure means there is less ambiguity for recipients, and more efficient test design for data handlers to address data quality issues early.

As messages carrying important data move between two systems, sometimes inside the same organization, this introduces risk of that data being incorrect or of poor quality. Consequently, even performing the same action between two separate entities from two external organizations can only further compound quality issues and security risks.

Two Key Ways Open Banking APIs Impact Data Quality

1. Protecting Outbound Data

Poor quality data flowing through the API will cause both operational and transactional issues through flawed reporting, searches and analytics.

Banking clerks or anyone in a customer -facing role, for example in a call centre, may have to search for specific customer accounts and, in the process, discover duplicated entries for the same individual. Duplicated records in online banking systems also cause confusion, misinformation and doubt – as well as taking up space in data lakes, costing the bank time and money. Time is wasted estimating the single point of truth for that account.

In the worst case, internal reports could be based on incorrect information. Imagine the potential impact of making strategic decisions at the C-Suite level on the development of products, or promotional rates to offer, based on inaccurate information.

2. API Security Challenges

Systems may be at greater risk of crashing when poor-quality data passes through APIs. Application quality assurance checks must be robustly designed to test against specific data structures and avoid security breaches or crashes.

Many instances of API security problems are completely unintentional. However, there are cyber-criminals who maliciously attack systems for personal gain. Leaving those organizations’ systems down for hours, days, and even weeks.

When exposing APIs to external parties, it is essential to ensure that security and data governance protocols are in place and data scientists can handle all kinds of data entities.

Centralized functions in financial institutions dedicated to data quality management practices must work with regulatory and legislation bodies to understand, gather and manage implementations under security certificates like ISO9001.

Solving Data Quality Challenges in Open Banking API Growth

Data quality is a fundamentally simple problem in open banking. The solution is often complicated. Failure to be compliant can be catastrophic.

While the volume and pace of technology growth is fast, it is challenging for organizations handling customer data to keep up with the changes. While tech is one of the world’s least regulated industries, banking is one of the world’s most heavily regulated industries and data quality is the key to security.

Customer data being shared with third parties through APIs will always present a risk. If any of that data is potentially inaccurate or incomplete, the exposure to risk increases.

To mitigate that risk, systems should have inbuilt augmented data quality solutions to monitor and assure the completeness and correctness of customer data as it arrives in the system from open banking APIs. This will flag up any issues which fall outside of the data quality standards being set.

In testing environments, using production-like data in simulated environments allows QA functions to mimic live data and present as many permutations of data as possible.

It is also judicious to utilize a data solution, like iData, which can obfuscate or synthesize data and present badly constructed data in order to be ready for all eventualities and protect against risk.

It is essential for synthetic datasets to have the same size as the dataset they are trying to imitate. This means that they must have a sample of all possible values, so that it can be prepared for any eventuality. The more prepared it is, the more likely the API will handle bad data and provide data certainty throughout organizations.

Adversely, it will be less likely to fail when presented with something unexpected.

Data quality assurance functions within financial institutions must have data governance practices and a centre of data quality excellence. By establishing these frameworks and consistent blueprints for data quality standards of open banking data, those problems can be solved much more rapidly.

This is a Sponsored Feature