At Funding Circle we want to help you earn attractive, stable returns by lending to small businesses across the UK. In the second installment in our series, we’re meeting with the teams that work behind the scenes to make that happen. Last month we spoke with the Credit Assessment team about how they assess individual applications. This time round caught up with James and Andor from the Risk team to discuss how they build our models and set our credit criteria.
Thanks for joining us. First up, what are your job titles and what areas do your teams look after?
James – I’m a Senior Data Scientist on the Decision Science team. We do the statistical modelling that sits behind our risk models. We’ll look at all the data we have on all our borrowers, including the information they provide when they apply, their financial statements and bank data, plus external data we get from credit bureaus.
Then we build statistical models so that we can identify and accurately price the businesses we can to lend to. This serves two key purposes:
- Identifies the businesses we don’t want to lend to
- Identify the businesses we do want to lend to, and give them an accurate price that’s fair to both them and to investors
Andor – I’m a Lead Analyst on the Credit Strategy team. Once we have the model from James’ team, we need to figure out what actions we want to take for each borrower type. For example, of the businesses we can serve, how much can we let them borrow? If they’re really low risk, they might be able to borrow more than others.
We’ll also look at what documents they need to provide. For example, we’ll often ask for extra information for large loans.
What goes into creating a risk model?
James – It starts out by identifying that we can improve our model. It might be that there’s a new data source available, for example this year we can start using CCDS data, which is bank statement data that comes from a credit bureau.
The next step is collecting all the data. This is usually the trickiest part, organising all the data into a format that will allow us to build models how we need to.
We’ll then start analysing the data. We’ll adjust different variables to see what impact they have on how loans perform. We’re looking to identify relationships within the data that can help us make better predictions on which loans will default or be repaid.
Once we’ve found those relationships, we’ll do extensive testing and validation of the model before passing it onto the Credit Strategy team.
Andor – We then use the model to make criteria for who to approve and what their interest rate should be. We’ll also look at how much they can borrow, what they need to provide when they apply and so on.
We then involve stakeholders from around the business to make sure everyone is aligned on the new changes. We’ll get feedback and provide training. We want to provide a framework for everyone involved so that we can get the best outcomes for all our customers.
We don’t just rely on models though. Our Credit Assessment team apply their experience to look at the individual circumstances of the businesses that are applying, then use the model, the criteria and all the information available to make a final decision.
How do you balance trying to give the best price for business and investors?
James – It’s a difficult balance to strike. If you wanted to minimise risk completely then you wouldn’t lend to anyone. Then at the other extreme, lending to everyone would have a very high level of risk.
Approving the wrong people or lending too much to a business is not only bad for investors, it’s bad for the business that then struggles to repay. We really believe in helping small businesses and want to help as many as we can, but we need to do it in the right way.
What expertise do you need in the team to do this?
James – The main role of our team is looking at lots of data and trying to get something useful out of it, so our team features a lot of PHDs in areas like physics, statistics, computer science and so on. You need to be passionate about finding relationships in data.
Andor – Our team has a really strong background in data analytics and credit risk. People have come from various large organisations and been involved in asset finance, credit cards, loans, and they bring all that knowledge to help us improve at Funding Circle.
Do you have other responsibilities?
Andor – We work with people from around the business and there’s always ideas popping up for ways we can improve things. If we can support on them we’ll always help out, so we get involved in ad hoc projects.
James – As well as the credit models, we also advise on fraud capture and marketing models. Similar to Andor’s team, we work with teams from around the business too. We have a lot of information that can be used to improve different aspects of the business, so we can look to improve things in a variety of ways.
What’s the culture on the team like?
James – Our team is naturally very analytical. We don’t like to take chances, if we have an idea we need to back it up, I guess we’re very critical in that sense. We’ll argue our corner, but it’s all about getting the best results for our customers.
Andor – I’ve worked in other large finance companies, and the big difference here is that you work much more closely with senior managers. Meeting with the CRO (Chief Risk Officer) elsewhere is rare and usually a big deal. At Funding Circle though, we can go and discuss ideas with Jerome (our Global CRO) or other senior leaders whenever we need to. Having access to their insights and experience in your daily work is really valuable and makes the whole team more productive.
How does it feel to work for a company that’s helping investors and small businesses?
Andor – What I think about is that we’re not only fulfilling a need for both sets of customers, but they go on to achieve even more afterwards. So whatever you do in the day to day, by helping small businesses grow or investors get a better return, that has an exponential impact on the wider economy.
James – My dad runs a small business, so on a personal level it’s a mission I really believe in. Most people work in a small business, so supporting them while also helping investors make a return is a really good place to be.
By lending to businesses your capital is at risk. Not covered by Financial Services Compensation Scheme. All information correct at time of publishing.