Before we get into the details of how risk assessments can be improved, let’s take a step back and start with some basics.
Generally speaking, a risk assessment is the process of identifying and analysing factors that could negatively impact assets – and making judgments accordingly. A common example is when lenders assess the risk profile of applicants to determine whether they should approve or deny their application for a loan. The goal, of course, is to ensure that they won't lend money that can't, or won't, be paid back.
Risk assessments aren’t just necessary for lenders, of course. Landlords and rental services, car leasers and telco companies, for example, also need to evaluate applicants and the potential risk of entering an agreement with them.
Since in most cases businesses are looking to understand whether a prospective customer might present a financial risk or not, they’ll make risk decisions by analysing the customer’s financial situation. To do that, lenders can request information from external credit bureaus, or request that applicants submit financial statements for them to review – that might mean proof of income (such as payslips and tax return statements), proof of expenses or information on loan exposure through (usually lengthy) application forms.
Most lenders use financial models to estimate the probability of someone’s default after borrowing. To feed and continuously improve these models, they need a wealth of data – and the more sophisticated and granular it is, the better. That’s where things start to get tricky.
Having a good understanding of someone’s risk profile has always been a challenge, because most data sources used in evaluations are outdated, costly and incomplete – meaning they can’t always provide the information needed to understand someone’s true risk value.
Plus, to remain relevant, today’s products and services don’t just need to be digitally available – they also need to meet high expectations for simplicity and functionality. Any application flow that requires significant effort from the user, or involves long waiting periods has a high likelihood of being abandoned.
To start with, a risk assessment process that gives your business more data points to base decisions on will, quite simply, help reduce your risk exposure while still improving acceptance rates. This means more business, at lower risk.
And speaking of more business – removing friction from the application flow with a speedy and convenient process that can give quick results will also mean your applicants will be more likely to stick around and seal the deal. There’s no big discouraging roadblock to make them abandon their application and go look for a loan elsewhere.Having a seamless user experience doesn’t just benefit consumers – it makes your business more attractive too.
So. By now you might be wondering what makes for a great risk assessment process. Here’s our take on that.
Let’s say a potential customer wants to borrow money to buy a property or subscribe for data services, and you need to understand their risk profile.
Ideally, you’d have an assessment process that meets their expectations (fast, simple, fully digital) while still addressing all your business needs (having enough information to make accurate risk decisions).
In this ideal world, your risk assessment process should then be convenient, reliable and complete. Here’s what we mean by that.
Everyone wants to get things fast, and if you can’t offer that, your would-be customers will be flocking to services that do. Having a speedy digital application process should be seen as essential.
And please note we’re talking about convenience in a ‘getting it solved with a few clicks’ way, and not a ‘submitting only 11 documents instead of 14’ way.Additionally, customers don’t want to wait around for days to get an answer – they want to know if it will be a ‘yes’ or ‘no’ right away.
Convenience isn’t just a benefit for users, of course. By being able to speed up and streamline your operations, as well as offer quicker results, you’ll also considerably improve business efficiency.
Reliability isn’t just a matter of ‘will it work, and can I trust it?’ – when it comes to assessing a loan, the information has to be 100% verifiable and secure. Beyond that, you also need to know that you can rely on the information to be up-to-date and a true reflection of the current circumstances of the applicant.
This might sound like a given, but the fact is many businesses still have to rely on months- or even years-old data (which is often what you get from credit bureaus).
Making risk decisions based on years-old data would be akin to going to a doctor and getting a prognosis based on your past years’ blood work. Results may not be as dependable as you’d hope.
With the ever-evolving labour market (think gig economy, decreasing lengths of tenure, furlough schemes) making fixed income streams not always be the norm, and the rise in the subscription economy meaning people may have significantly less disposable income, you really need more granular information to have the full picture of someone’s financial standing.
With recent advances in data processing and risk modelling, when it comes to applicant information – the more data you can get, the better.
You should look for more than information on what the applicant’s loan exposure was six months ago. The more you can closely examine someone’s spending patterns – and how they’ve changed over time, the better you can identify markers that could indicate ‘risky behaviours’ such as gambling, or reliance on overdraft.
As mentioned, the common methods used to determine an applicant’s creditworthiness – reports from credit bureaus, documentation submitted by the applicant (or in some cases internal data from the financial institution they’re applying to) – have their shortcomings, and may not always reveal the applicant’s true risk.
To start with, credit reports are limited to the applicant’s loan exposure. So, if they pay back their loans on time, you’ll see that reflected in the reports. However, credit reports won’t indicate what their account balance is on the first and the last day of the month (maybe they’re really struggling to get by) or whether the applicant may have made several dodgy ATM withdrawals in the past few days.
As for requesting documentation from the applicant – it doesn’t score any points for ‘convenience’ and the results won’t always be reliable either. It comes with a lot of friction for the applicant, not to mention high operational costs of gathering and processing the information. And even though everything may seem legit – financial statements can be easily falsified.
Finally, internal data can be a valuable source of information, but it will be limited to your existing customers. While the data is reliable and can be reached instantly, it might not always paint the full picture. After all, people can have multiple accounts with different providers.
So what can you do when you’d like a risk assessment method that does tick all the boxes?
By providing a way to access consumers’ bank account data, open banking is making it easier to perform more in-depth risk analyses. With fresh, real-time transaction data coming straight from the applicant’s bank, the information won’t just be reliable, it will also be reliably up-to-date.
You also won’t just get data from the past few months (say, as you might if asking for a couple of recent payslips) – the data can go back for years, and it gives a very deep understanding of the applicant’s financial capability and spending behaviours. This makes it more complete than most other methods out there.
Finally, open banking makes the process convenient, as it doesn’t require any real effort from the applicant – or the lender. The applicant just needs to authenticate with their bank to authorise the use of their data (which can be done with a few clicks). And the lender doesn’t need to go through mountains of data to make sense of what’s going on – at least not if they’re using a solution that makes sense of the data for them. Like Tink’s.
With Tink’s Risk Insights, the data digging is done automatically in the background, resulting in an in-depth risk analysis highlighting different risk factors for lenders to consider. It works like an MRI scan of someone’s finances, revealing risk patterns that might not be spotted by credit reports – such as gambling, ATM behaviour, frequency of overdraft, and more. And, importantly – it shows these risky patterns historically, so you can also know how these behaviours might have changed over time.
Here’s how the user’s application process might look like when doing a risk assessment with Tink (scroll through the images below) :
