Open data can now be categorised to bring value that we have not seen until now

“To extract value from open data, you need to categorise it. When you do, the results are huge”

Open Banking is the next generation of data sharing . But it’s not new. Transactional data sharing has been around for several years in the guise of screen scraping and is already used by more than 2 million people in the UK. In the business world, the Mandatory Credit Data Sharing regulation has the same premise; data sharing of commercial transactions .

What Open Banking is doing is providing a stronger governance, security and consistency in the supply of data.

Regulatory initiatives such as the General Data Protection Regulation (GDPR) strengthen the requirements to share data in a secure, compliant and auditable way. More importantly they will be tasked with helping customers understand the value of their data.

The most recent FCA consultation implies that organisations will need to look again at the tools used to measure a customer’s credit worthiness, notably their credit and affordability risk. These tools need to focus on understanding an individual’s financial status but must be used proportionately in relation to the credit product being sold and the associated risk.

Consideration is also starting to focus on understanding the risk to both the lender and consumer throughout the life of the loan.

In response, many lenders are looking to change their methods for assessing affordability based on the ‘real-time’ analysis of Open Banking transactional data to inform income and expenditure.

It’s not uncommon to hear people ask whether the information provided by Open Banking bring an end to credit scores. Regardless of our business, we don’t believe it will. Why? Because credit scores inform different parts of a customer’s financial behaviour. Credit scores provide a measure of an individual’s credit risk – their credit history and willingness to repay the credit debt. Affordability ‘risk’ considers an individual’s disposable income and capacity to afford a loan, now and in the future. These two data points complement, rather than compete against each other, providing a more sophisticated view of financial well-being.

More and more we’re seeing a combination of Data Science, Machine Learning and Artificial Intelligence , being introduced to analyse this data and automate decisions. We believe this will revolutionise the way people interpret data and change the way credit decisions are made. Machine Learning will speed up the process by which you can inform your understanding of an individual’s creditworthiness and what they can afford. This has provided a big opportunity to reduce the time taken to assess an individual’s affordability and the associated costs of doing so. More importantly, if used correctly, it can relieve friction in the customer journey . Open Banking gives people the opportunity to consent to share their data instantly, across lenders, to inform a decision.

Take mortgage applications. Open Banking is set to revolutionise this market by allowing consumers to share their bank statement data in real-time rather than relying on paper-based proofs of bank statements. In an open data world , this paper trail can be automated. It’s instant, accurate and secure. It eliminates the need for manual interpretation of a bank statement by an underwriter and enhances efficiencies that we, until now, have not been able realise.

Many lenders are already looking at the role of transactional data to pre-qualify a customer’s eligibility for a product. In tandem, they are also analysing its ability to help manage customers throughout the life of their loan. For example, if you can get access to the data with prior consent from the customer, you could use it to proactively manage that customer as their financial circumstances change. For example, for predelinquency. It can also highlight those customers who have more headroom and the propensity and appetite for, additional products.

People’s financial circumstances change on a regular basis. If we get the value exchange right, and persuade them to share data, then we can respond and react. We can deliver ‘real’ customer value. APRs and credit limits can be personalised accordingly, managed in response to a customer’s financial well-being. You could better support people who are currently not able to access mainstream credit – and those who have thin files (like new to country residents or young people). You can better understand the income deposits of a self-employed person through categorising their deposits, like never before. Opening a significant opportunity for you, and them. This is where transactional data sharing could be important and very rewarding.

Applying a real-time categorisation engine means transactional data can be categorised, aggregated, summarised and used to supplement a traditional credit score. At Experian, through investments in the areas of open data and Open Banking , we can now categorise 12 months’ worth of bank transactional data in under a second. Outputs from this engine, – named Trusso – can be used in association with automated decision software to optimise credit decisions and reduce the time taken to offer, react or respond.

As with everything now, it also has flexibility of scale and agility to be aligned to your own lending rules and used to predict outcomes based on your organisations current credit risk policies. It can also help you to create new lending rules based on a more accurate and personalised view of your customers .

This architecture, which we have invested in and tested, is available now. The next phase is to embed this into the financial ecosystem and realise the benefits it is poised to deliver.