What does your bank account say about you?

risk blog

Have you ever considered what your bank account would say if it could talk?

Since its creation in the 80’s the online bank account has taken up permanent residence in our lives. We use it to pay bills, transfer money to friends and family, and most of us, use it to track our spending.

As a result, our online bank account is a large, wholly unstructured dataset packed full of important insights about our spending and lifestyle habits. The same can be said for business bank accounts, which is what we commonly deal with at Spotcap. Major businesses, and fintech players have cottoned onto this, but are still learning how best to leverage it.

In order for raw data, like the seemingly meaningless numbers and letters contained in our bank accounts, to provide meaningful insights, a data scientist must first work their magic. A study conducted by Crowdflower last year found data scientists spend 60 percent of their time interpreting, cleaning and organising data.

Business bank account data, when structured correctly, allows us to identify a business owner with a handle on inflows, outflows and working capital. We can see if a business relies too heavily on one client, or if the success of a business is too seasonal to be sustainable. Essentially our machine learning models can make a prediction about how a business will perform over the next six to 12 months.  

Bank account data gives real time insights

It’s my job as Spotcap’s Senior Risk Analytics Manager to construct the data found in a bank account into a form of variables enabling us to derive meaningful insights. We use an experimental process and constitute a wide range of numerical and categorical attributes built both from transaction amounts and their descriptions. With the right set of variables in place, a bank account can paint an interesting, yet somewhat abstract, picture of who we are and how we live.

With that in mind, it’s easy to see why fintech lenders frequently turn to bank account data as an integral component of the risk assessment process. While we gather insights from business activity statements, profit and loss statements, balance sheets and tax returns, these sources take a backward looking view, as they rely on an aged data set. We can derive far greater insights about business performance from bank account data, the most current source of information at our disposal.

You cannot hide anything in a bank account

We also use the data to validate and crosscheck the information gathered from traditional sources. Why? Because it’s impossible to hide anything in a bank account.

It’s the most trustworthy, reliable, verifiable and up-to-date source of business intelligence we have at our disposal. It enables us to make robust lending decisions in a fast and efficient way.

At this point, you may be wondering why, and how, we have access to our client’s bank account data. It’s a fair question. We have been conditioned to keep our bank accounts to ourselves. The truth is, our clients grant us one-time, read-only, access to their bank accounts for the sole purpose of completing our credit assessment. The data provided by our clients is encrypted using one of the strongest security systems commercially available, and we don’t store this information, even in its encrypted form.

Safety and security

In today’s digital economy, there is a lot (think tailored financial advice and products as well as a more holistic view of your finances) to be gained by giving fintechs one-time, read-only access to your bank data.

So much so that new EU wide legislation, The Payment Service Directive 2 (PSD2), which comes into effect next year, will ‘encourage’ banks to open up data to trusted third-parties, as long as you, the customer, ask for it to happen.

The directive makes an important and interesting statement about data ownership. Regulators believe the customer owns their own data and should therefore have the right, and the opportunity, for their data to be used more effectively by fintechs and other players. What do you think?

To find out more about Spotcap’s innovative approach to risk assessment check out our website.

konrad

Konrad Semsch is the Senior Risk Analytics Manager at Spotcap, a multinational fintech lending business. Konrad has played a key role in designing and building Spotcap’s end-to-end scoring infrastructure and machine learning scoring models. The bank account text mining tools Konrad has written have enhanced Spotcap’s proprietary credit algorithm ensuring a robust credit assessment process. Konrad brings more than five years of experience in building reporting tools and analytic solutions to Spotcap, having previously worked for Rocket Internet, Unilever and Diageo. Konrad holds both Bachelor’s and Master’s degrees in Quantitative Methods in Economics and Information Systems, from the Warsaw School of Economics as well as a Master’s in International Management from the University of Sydney, Australia.