Sound Businesses: Profiles of companies and business models we are keeping an eye on.
U.K.-based digital bank OakNorth Holdings Ltd. received a total of $440 million last month, including a $390 million investment from the Vision Fund of Japan’s SoftBank Group Corp. and the rest from the Singapore-based Clermont Group.
Only two years old and already profitable, OakNorth has carved out a unique path by making much more consequential loans to small and medium enterprises, compared to players like Ant Financial Services Group, PayPal Credit or Kabbage.
What makes OakNorth’s offering compelling to other banks is Acorn, its cloud-based technology that allows banks to gather data and insights about small businesses. OakNorth’s latest round leaves it at a $2.8 billion valuation, making it Europe’s most valuable technology startup.
Valentina Kristensen, the company’s Director of Growth and Communications, spoke to Karma Network’s Contributing Editor Michael Moran about OakNorth’s plans for expansion in the United States and its mid-market approach that has won over larger global commercial banks.
Michael Moran: OakNorth is profitable after only two years. That would stand out in any crowd, obviously SoftBank and other investors agree. What accounts for OakNorth’s distinct competitive advantage in the crowded digital banking space, and how will that now be applied to the US market?
Valentina Kristensen: Yes, it’s a pretty crowded space and besides our liquidity, we have a good idea of OakNorth’s competitive advantage. To make it very clear, we’re not coming to the United States to replicate what we’re doing in the United Kingdom. For instance, we’re not seeking a U.S. banking license and in the U.S. We won’t be doing our own balance sheet lending. The plan for success is two-fold.
First of all, by continuing to license the OakNorth platform to other banks and lending institutions and that will enable them to do the underwriting, credit, underwriting and analysis of loans and also the private debt monitoring of their portfolios.
And secondly, in the U.S. we will take that a step further by actually partnering with banks as well on the origination of loans. So we will help them to originate, underwrite and monitor loans. But the actual balance sheet lending will come from a U.S. bank that has a U.S. banking license.
So again, to be clear, in the U.K. we are a fully licensed bank. We have our inbound bank license which enables us to raise retail deposits and then use those deposits to help fund our lending and we’ll continue doing that. But outside of the U.K., we do not have plans to seek a banking license. We’ll be partnering with banks in the U.S. and then they’re doing the actual lending from their own balance sheet.
Michael Moran: What do you bring to a traditional bank that they could not do internally themselves?
Kristensen: Our biggest competitive advantage is speed and that’s obviously where the technology is hugely impactful. You know, the fact that we can come back from the first meeting to cash disbursement in weeks rather than months means that we’re saving a huge amount of time, money and resources compared to larger banks.
Customers who are coming to us for a loan have a much better experience, because they’re able to avoid that opportunity cost they would get from a traditional bank and get a quick yes or no decision. Typically, if it’s a ‘no’ we can tell them in a matter of days.
Michael Moran: How is this different from what existing peer-to-peer lenders are doing in the U.S. and elsewhere? Do you view them as competitors?
Kristensen: Well, they typically deal in tens of thousands of dollars or perhaps up to a couple of hundred thousand dollars. As I mentioned earlier, you’ve got the “Kabbages”, the SoFis and such, and they typically base decisions on a credit score model. And the machine essentially tells you, “Yes, we’re going to go ahead with the loan,” or not. So it operates in a similar way to taking out a credit card loan, for example, and there’s a lot of data available in the market on that.
What you don’t have is the sort of next stage of lending which is this mid-market, SME space, where you’re talking about loan of half-million dollars up to tens of millions of dollars. There you almost need the bespoke approach that you find at the institutional level, but can’t access it.
You need it with the speed that you’d find in the peer-to-peer space. Our ultimate goal is to get to that sort of timeframe, to where it’s days, versus weeks, for a decision. Even now, our approach is weeks vs. months, and that’s already significantly better.
Michael Moran: In the absence of the kind of corporate SME credit rating system, which underpins the peer-to-peer lending model, how do you do credit risk and other underwriting calculations?
Kristensen: So the platform doesn’t work by assigning a credit score or providing us with a yes or no. It’s about the machine enhancing the human rather than replacing the human.
The platform works by parsing millions of data items on SMEs from different sectors, parameters, and geographies, and then uses machine learning algorithms to pull out the data and put together a 40-page document that the credit committee use to make a decision.
It’s still a human making decision at the end of the day, so in terms of the KYC and compliance, that’s all still part of the system during the credit analysis and client onboarding stage.
Michael Moran: The typical evolution of digital banking has been to start with a payments system and then start lending based either on credit scores or, particularly in emerging markets, on data derived from the payments relationship. But you’ve skipped that and jumped right into the SME market. Can you explain the thinking behind that?
Kristensen: This was this was basically one of the main drivers, or one of the main drivers, behind SoftBank’s interest.
Reporters have been asking if this SME focus and the licensing of our software to larger, legacy banks was a differentiator, and why have we never heard of it until now. Why we didn’t come out when we first came to market and scream and shout about that?
Our founders, Rishi Khosla and Joel Perlman, had the view that we didn’t want to do the typical route of building a technology and then going out to banks and telling them we’ve got the fantastic technology to solve this problem finally at the mid-market lending level.
Instead, we went out and used it commercially. We decided we’d much rather prove it, you know, prove that we can actually do what we say we think we can do with this technology. In the U.K., which is a highly regulated and highly competitive market, if we do manage to prove it, then at that point we can you know take it to market and say, “okay we’re ready to license this out to other players.”
Michael Moran: Who are your competitors?
Kristensen: Despite (SME lending) being a $7 trillion market globally, at this particular level, you know that there hasn’t been a competitor in our space. It’s not that there aren’t numerous players like you see at the lower end of the market. But there isn’t that much data available for SMEs and it’s a lot more sensitive credit scoring SMEs than doing it based on credit scores for loans of a couple of hundred thousand dollars.
But when you get to this next level, it’s difficult for a couple of reasons. One, because obviously you know you don’t have as much data available as you have at the lower end. Two, because you can’t just take the same approach that you do at the lower end of the market because each business applicant is different.
You know the assets that each business will have will be different and you need to take a more bespoke approach to underwriting. You wouldn’t have to go into that level of rigorous underwriting at the lower [peer-to-peer] end. It’s almost more akin to what you’d find in a private equity firm rather than what you’d find from a traditional high street bank.
Meanwhile, traditional banks are only willing to look at real estate as a means of securing loans. They’re not willing to look at the assets of a business like stock, debts, plant machinery, intellectual property, alternative assets like art. A great example of this is a borrower we have which is a gallery on Bond Street (London). As part of their security pool, we’ve got security against four pieces of art in the gallery. That’s the kind of bespoke approach that you simply don’t find typically from a traditional high street bank. That bank is very happy to get to know a business if the loans in question is several hundred million or even billions of dollars.
When you are a regulated bank, you can’t automate it like you can for much smaller sized loans. That’s been the problem historically and that’s why it takes several months to get a decision. So we’re saying, “Well, technology can take away that manual labor and then, in effect, reduce the time frame, so this no longer becomes an issue for the bank and it should no longer take several months to turn back loans
Michael Moran: Can you take me through the company’s founding story? Where did the original idea to focus on SMEs come from?
Kristensen: OakNorth was founded by our CEO Rishi Khosla and his partner Joel Perlman. They met at (London School of Economics) while studying for their masters and started a business in 2002 in their late twenties. That business, Copal Amba, outsourced financial research with big data and credit analysis for investment banks, asset managers and private equity firms. Over a period of 12 years, they built it up to a 3,000-person business and sold it to Moody’s Analytics in 2014.
They had bootstrapped the business from the very beginning from 2002 with about $60,000 cash and in 2006 they were looking for finance to scale the business. They were profitable, had very strong cash flow and a great list of clients, so didn’t want to go down the equity route and dilute. They felt that they had a good enough model and a prudent position that they should be able to get debt financing. But because in 2006 there was no peer-to-peer lending – no platforms like Kabbage, for instance – their only option was to go to a traditional bank.
But the traditional banks only lend them if they could secure the loan against a property and the maximum amount they were willing to lend them was $130,599. Eventually they would up at Citibank’s “Special Situations Desk,” and after some convincing they managed to get Copal Amba a $10 million dividend recap.
So that was the premise behind OakNorth. The idea that, if you look at the market today as a startup, it’s typically pretty difficult to get equity financing because you haven’t got a proven business model. You don’t have a track record or history and people really have to kind of put their faith in you as an entrepreneur.
But once you get to the sort of scale up stage where you grow a profitable business, you’ve got a proven business model, finding equity is actually pretty easy. You know lots of people want to throw cash at you, the challenge then becomes actually finding debt and it’s not that banks won’t lend to you. It’s just that typically at this size of the market, a loan of half million dollars up to $20-30 million, it’s going to be a very time-consuming and typically quite a painful process to get that loan. And even then, traditional banks are not willing to structure a bespoke package to suit your needs. They’re not going to look at the other assets of a new business; they’re just going to look at real estate. If you don’t have any, then they’re not going to go ahead with the loan.
Let’s say you’re a restaurant chain owner or a hotel chain owner, and you’re trying to find finance for a scale out of your business. You apply, go into due diligence and then you have to wait nine months to get an answer from the bank as to whether or not you can actually get the finance. Imaging then you don’t get the financing.
That’s a painful process we’ve now got down to a few weeks, and we are hoping to get it down to days eventually.