The analysis begins with the first keystrokes of a borrower looking for a loan on Kreditech’s site, collecting clues about personality and identity – for example if someone types in their name or glues it.
Copy and paste a name suggests a scam bot rather than a human. Someone desperate for money is likely to jump straight to the registration page rather than spending time reading the website. Potential borrowers are encouraged to share their browsing history, giving the business access to their purchasing habits and whether they spend time on websites that could make them a credit risk, such as gambling. money online.
Kreditech, a Hamburg-based start-up that offers credit and banking products to people with little or no credit history, has caught the attention of investors such as JC Flowers and Peter Thiel. The vision of the fintech company is to create an “Amazon for consumer credit” that expands to offer personal finance management products, overdrafts and payment services.
Traditional credit scoring looks to the past and the present as a guide for the future. But what if many data points, from psychometric tests to posts on Facebook or LinkedIn, could be used to paint a more accurate picture of creditworthiness?
Rene Griemens, CFO of Kreditech, says: “Historical behavior can be an indicator for the future, but for most people it is not, because for most people a better indicator is their. current behavior and their probable evolution. “
Instead of focusing on the repayment history, Kreditech seeks to uncover the “underlying personality” of a borrower from a richer data set. The company consults information from social networks, which is voluntarily shared by a loan applicant.
“Social networks tell you a lot about the person,” says Griemens, at Kreditech’s office in a building with a stunning view of Hamburg’s waterfront.
“What kind of friends do you have? We may be able to see if he has any friends who have already paid us off a loan – this is usually a good indicator. “
Social media gives lenders insight into how an applicant spends their time. A borrower who lives in St. Petersburg but spends all of his time in bars in Moscow may be at high risk, as his location suggests that he spends more time partying than working. Another tool in development analyzes the sophistication of the language a user uses in Facebook posts or tweets. However, if a person refuses to share their profile on social media, Kreditech will not automatically disqualify them.
It’s not just the sheer volume of data that differentiates online lenders such as Kreditech; the company is also seeking to transform the classic definition of solvency.
Rather than ticking off borrowers against a checklist of requirements like minimum income and age, the company lets its algorithm map an applicant’s behavior. Looking for correlations in a multitude of data means that there may not be a single “right answer”. Behavior that might be suspicious of a factory worker – applying for a loan in Hamburg while his job is in Stuttgart – might make perfect sense for someone who has an itinerant job.
Establishing behavioral models helps a lender look beyond past defaults; a borrower may have tried to repay a previous loan, but had difficulty doing so on a low income.
Mr. Griemens recognizes that even algorithms can be faulty. “The algorithm is as good as the dataset it was trained on,” he says.
Kreditech is too small to worry the big banks. Its loan portfolio is in the “average double-digit millions of euros,” Griemens said. The company, which is in deficit, estimates its turnover for 2015 at around € 41 million.
But the opportunity for a business that obtains this right is significant. Look at a shaded map of the world by credit scores and there are large empty spaces. In many countries, large sections of the population are not credited. Even in countries where every adult has a credit score, a more accurate representation of creditworthiness could open up untapped segments of the market.
Although based in Germany, Kreditech’s main markets are the Czech Republic, Mexico, Poland, Spain and Russia. In some of these markets, the start-up grants loans to people who otherwise could resort to the gray market for informal loans. The company estimates the total size of the global consumer credit market to be around $ 8 billion.
Like others who extend funding to borrowers at the margin, Kreditech’s rates are high. The company charges a standard interest rate of 0.8 to 1% per day on a microcredit, which is repayable over 30 days.
The company was founded in 2012 by Alexander Graubner-Müller and Sebastian Diemer, both of whom had worked at German tech start-up conglomerate Rocket Internet. While at Rocket, the pair looked for a partner to provide financial services to e-commerce customers in emerging markets, but struggled to find a solution.
Mr. Graubner-Müller, who took over as Managing Director in November and Mr. Diemer to focus on a longer-term strategy, says: “Banks did not have the capacity to assess this type of customer. . . or the credit application processes were not designed to be user-friendly in an online context.
The new CEO looks restless and slightly impatient of a bright schoolboy, with blonde hair and black rimmed glasses. Mr. Griemens, whose career spans two decades, has a more relaxed and affable manner. He joined the company in 2013 and his CV includes stints at Citibank in London, McKinsey in Portugal and more recently at a boutique bank in Berlin.
The location in Hamburg for the company headquarters was due to chance. Both founders were based there at the time and found angel investors locally. The city’s beautiful brick architecture and lively clubs and bars make it an attractive place for staff. Kreditech employs around 200 people worldwide.
Considering the number of people around the world who don’t have a credit score, it’s no surprise that Kreditech isn’t the only company trying to apply new technology to assess a borrower’s reliability.
However, Oliwia Berdak, analyst at Forrester, warns that innovative credit scoring systems have not yet come under pressure. She says, “The problem is, this is just the beginning and they haven’t gone through a financial crisis that would put a lot of pressure on these loans.
To this, Mr. Griemens says: “Our system has generally worked well in negative macroeconomic environments; of course he has to ‘learn’ how to handle such environments, but overall we would expect the models to be able to adapt faster and better than loan officers did in the last 2 years. Last 000 years of banking.
* This article has been modified to correct a fact regarding the Avant review process