© The Financial Times Ltd 2016
FT and 'Financial Times' are trademarks of The Financial Times Ltd.
The Financial Times and its journalism are subject to a self-regulation regime under the FT Editorial Code of Practice.
December 13, 2012 6:30 pm
The exposed brick walls, open work benches and large flat screens in the San Francisco office of Sasha Orloff’s start-up make it feel like any other trendy young Californian internet company. But Mr Orloff is not tinkering with the photo-sharing apps or other consumer trivia that litter the start-up landscape: his company, LendUp, has just joined the payday loans business.
A practice at the bottom of the lending scale that turns on advancing unsecured money to people with little or no credit standing, this is an industry whose more familiar public face is a ramshackle store with large dollar signs over the door.
Some of Silicon Valley’s brightest think they have a better approach. By logging in with their Facebook details, LendUp’s customers have the option of exposing their online social behaviour to its data-crunching algorithms. Watching how you behave online with your friends, according to the company, is a good indicator of how likely you are to repay a loan.
“The hypothesis is that strong social ties and community are a benchmark for responsibility,” says Mr Orloff, a former credit analyst at Citibank. “Are you in your community, or are you dabbling here and there?” The stronger and more active your social connections, the more likely you are to get a loan.
As new sources of data multiply and the cost of collecting and processing it collapses, ideas like this are springing up all over the financial landscape. With its core products founded on information, the financial industry is more open to disruption from heavy-duty data-crunching techniques than most, according to investors who have flocked to the business.
“Everything [in finance] really springs from data,” says Roger Ehrenberg, a former hedge fund manager at Deutsche Bank who now makes investments in so-called big data companies.
The falling costs of computing and communications lie behind the explosion in data that is triggering this upheaval. The founder of Progressive Insurance, a US car insurer, first dreamt of collecting information about his customers’ individual driving habits to help with risk assessment in the late 1980s. It was only in 2010 that the idea became cost-effective: since then, more than 1m customers have installed devices in their cars that report back how abruptly they brake or whether they are heavy on the accelerator.
“The network of sensors is opening a universe of data that previously you couldn’t get at,” says Richard Hutchinson, general manager of usage-based insurance at Progressive.
The architecture of the open web has also contributed, making it easier for newcomers to ingest feeds of digital information. Drawing on APIs, or open interfaces, from companies such as eBay and PayPal, Kabbage, an online lender to small businesses in the US, says it can track “thousands of data elements” that help reveal the health of a customer’s business. Through a special deal with UPS it also gets parcel shipment data. Throw in Facebook, and it has become possible to get a detailed sense of how a small business is faring,says Kathryn Petralia, chief operating officer: “You can measure the level of reciprocal engagement between a company and its customers.”
In the consumer field, strict laws in the US limit the use of data like this to make credit decisions that might hurt particular groups of individuals. But that hasn’t stopped the experimentation. Besides information from traditional credit-rating bureaus and social networks, LendUp mashes up many other public sources: if a customer has changed cellphone numbers multiple times, for instance, it is probably an indicator of weak social ties and hence a bad credit risk, says Mr Orloff.
Number crunching on an epic scale is not new to consumer finance. The credit card industry has been among the most advanced when it comes to using data for market segmentation and database marketing. Nor is there anything unusual about the idea of an interloper using superior data analysis techniques to carve out a significant slice of an established financial market: CapitalOne managed the feat in the 1990s, becoming one of the largest US credit card issuers.
The new technologies of big data, however, are threatening to level the playing field with even the most data-intensive incumbents. Companies like LendUp and Climate Corporation, which analyses weather patterns to price crop insurance for farmers, say they run frequent massive simulations with their data, using technology platforms that were purpose-built for heavy-duty number crunching. Rivals with legacy platforms may not be able to respond as quickly.
The technologies of the cloud have also lowered the economic barriers to competition, says Zach Bogue, a founder of Data Collective, another group that invests in big data companies. Rather than facing significant upfront capital spending, start-ups can now tap into the processing power of companies like Amazon on a pay-as-you-go basis, he says.
Core disciplines like risk management, credit assessment and consumer marketing are all being influenced by this rise of data-intensive applications. But the next frontier stands to be more disruptive, as big data usher in a new level of product differentiation.
Dynamic pricing – or using real time data to price services on the fly – points to one of the areas of greatest potential, according to many investors and entrepreneurs.
For instance, Wealthfront, an online investment manager, says it adjusts its clients’ holdings frequently to recognise losses that bring tax benefits. The technique, known as tax-loss harvesting, is usually only practised at periodic intervals: turning the practice into a dynamic activity stands to add 1 per cent to investment returns, according to Andy Rachleff, chief executive.
Progressive is also looking at ways of applying real-time data to create better services, says Mr Hutchinson. “When you start capturing data on a per-second basis, the expansion of the data is immense.”
Among the new ideas are utility-style pricing, under which drivers would be charged based on how often they use their vehicles, and insurance plans for car-sharing services that adjust the policy depending on who is using a car. As always in finance, such innovations are likely to come first from the most forward-looking companies with the best number-crunching skills. The challenge for the rest, as the pace of innovation accelerates, will be to keep up.
Copyright The Financial Times Limited 2016. You may share using our article tools.
Please don't cut articles from FT.com and redistribute by email or post to the web.
Sign up for email briefings to stay up to date on topics you are interested in