In your editor’s opinion it will help the underbanked if privacy advocates and data protection fanatics will let them. 

ZestFinance is just another example of game changing data mining and analytical tools permitting lenders to bypass traditional credit information providers or credit bureaus.  ZestFinance is one of a fast-growing coterie of financial technology companies – “fintech” that are aiming to revolutionize the way banks and other financial services companies operate.   The technology will not only upset privacy advocates and data protection commissioners but also the traditional and well entrenched decision analytics companies such as CRIF, Experian, Equifax, FICO and TransUnion.

The Financial Times recently published an article about zestFinance.  The article was sent to me by a colleague from the credit bureau industry.   He was clearly worried about the consequences of ZestFinance type developments on his business.  In my opinion he should be.  Rather than taking the lead in big data/social data mining, credit bureaus in Europe, for example, caved in to negative public opinion, rather than promoting the benefits in providing much needed access to finance for the underbanked, micro businesses and small businesses.

The article of the Financial Times comments on the development of fintech and the rising star ZestFinance.  Talk of fintech investment is all the rage these days, as banks try desperately to focus attention on a positive forward-looking agenda, and draw a line under the legacy woes. This is partly spin, of course.  But investment in technology is also genuinely needed after seven years of crisis and post-crisis regulatory reform that distracted focus – and funding – away from the pipes and cables that make everything work.  Who knows? Done right, it could even help prevent some of the scandals of the past. And, as with so much in the mass market internet world, it can claim to be an agent of democratisation.  Zest and companies like it claim that many people who have traditionally been denied credit – either because they lack a record of previous borrowing, or because that record is bad – will be validated by their broader approach to credit scoring.

By using “big data” records sourced from individuals’ social network and internet footprints, typical credit scores can end up 40 per cent higher, Zest says.

That could be good for the lenders, who can take on more business with less risk of default, and good for the borrowers, who want the money. This data analysis can be particularly powerful in emerging markets, where families may need more money than they have to hand to school their children or start up small businesses – both vital investments for macroeconomic development. Admittedly, some of the data mining that fintech companies carry out makes liberty campaigners jittery.  On balance, though, it is refreshing to see new ideas being brought to the traditionally obscure process of granting consumer credit. And if big data gives the industry a second leg to stand on, a third innovation is emerging from another oblique angle – and appears equally promising. A Shoreditch business called VisualDNA has been gaining rapid traction for its credit scoring of loan applicants through a form of psychometric testing.

“Not everyone has a credit score, but everyone has a personality type,” argues Alex Willcock, the company’s founder. VisualDNA’s test, administered via a lender, is relatively quick, taking 10 or 15 minutes. The bulk of the questions ask applicants to choose pictures from a selection that can reveal their attitude to types of risk.  For example, one offers pictures of a preferred leisure activity: a day out with the family, a night out partying, or a good book? The company’s technology has been up and running for a couple of years in Russia, via Svyaznoy, a phone retailer-cum-bank that is controlled by VisualDNA investor Maxim Nogotkov. Among the dozen or so lenders that now use the tests across half a dozen countries, some report improvements of as much as 50 per cent in loan default rates. As a result, credit scoring agency Experian and credit card company MasterCard both signed up as clients, and banks in South Africa, Turkey and Mexico are about to trial the tests.

Mr Willcock is not lacking ambition. He talks about targeting 30m borrowers out of a 1.9bn “addressable unbanked” market over the next few years. There are some practical shortcomings – some pictures are open to interpretation, and some applicants may dodge safeguards designed to spot dishonest answers. At a macro level, too, there will be concerns. Improperly used, this kind of “fintech” could further inflate the credit bubbles in many markets to unsustainable levels. But, like it or not, technology is the future. The world is not going back to the old order of a bank manager knowing each of his customers individually and making credit judgments on that basis. Big data and psychometric tests have the potential to replicate the personal touch, and ensure lenders know more about their customers than they ever used to. In recent years, amid money laundering scandals, the bank maxim of “know your customer” has become all about regulatory compliance. But it should also be about making decent business decisions.

Source:  Financial Times

About ZestFinance:  ZestFinance was founded by a team of (mostly) ex-Google and Capital One folks who came together with a mission – to save the underbanked billions of dollars. Since then, the team has grown to more than a hundred crazy smart data geeks, mathematicians and computer scientists, all working together to re-invent underwriting and make more credit available to the people who need it most.

What Does ZestFinance Do?  For the past 40 years, traditional lenders have used virtually the same underwriting methods to decide whether to offer credit. These methods assume that it’s hard to get a lot of data, it’s expensive to store it, and nearly impossible to use huge volumes of data.

The era of big data has changed all that.

Google-style machine learning meets Capital One-style credit scoring.

Traditionally, underwriting has used a handful of variables. ZestFinance takes an entirely different approach to underwriting using machine learning and large-scale big data analysis.

With a team of some of the world’s best data scientists from Google and lending experts from Capital One, ZestFinance analyzes thousands of potential credit variables – everything from financial information to technology usage – to better assess factors like the potential for fraud, the risk of default, and the viability of a long-term customer relationship.

Higher approval rates, lower default

Our big data underwriting model provides a 40% improvement over the current best-in-class industry score. That translates into more accurate credit decisions, which leads to increased credit availability for borrowers and higher repayment rates for lenders.

This new approach to underwriting will enable lenders to expand their customer base, take business from their competitors, and better serve existing borrowers – without affecting their default rate. Alternatively, lenders can use our model to lower default rates while maintaining a particular approval rate, limiting losses and significantly improving returns.

Bottom line: Big data underwriting will save you money. Lots of it.

ZestFinance Can Help:  ZestFinance helps lenders in all credit segments better assess the credit risk of potential borrowers. Its technology can supplement or replace an organization’s current underwriting algorithms. With a proven, transparent method to help lenders measure the impact of our underwriting platform against current models, it can also help you assess results in advance.

How Does Zest Do It?

When founder Douglas Merrill (ZestFinance CEO and former Google CIO) first encountered the challenge of scoring a subprime population, he immediately thought of Google’s high-dimensional machine learning search algorithms. This technology is optimized to handle missing and changing data because it consumes thousands – if not hundreds of thousands – of variables. He believed these same algorithms could work for consumer credit – and they can.

Traditional Underwriting Models are Limited

Everyone uses the same methods to underwrite. They either use logistic regression, decision trees, or a combination of both. Logistic regression can only handle 10 – 15 variables; additionally, all variables must be present (and correct). Decision trees require it be possible to divide all applicants into mutually exclusive categories.

Neither of these constraints really make sense. And when either is violated, the decision is incorrect. This isn’t surprising — everyone knows these models are imperfect. However, they are not only imperfect, they yield incorrect answers, especially for the underbanked.

ZestFinance’s big data approach avoids these pitfalls.

All Data is Credit Data

ZestFinance uses machine learning techniques and large-scale data analysis to consume vast amounts of data and make more accurate credit decisions. When you analyze trends across thousands of signals, your ability to underwrite massively improves.

The ZestFinance decisioning infrastructure runs dozens of individual underwriting models in parallel and returns underwriting decisions in moments.

The Math Matters

The math matters in big data. If you get the math wrong, the data is useless or completely misleading. We get it right. ZestFinance has combined the best data modelers in the world with the best credit analysts in the field to build a big data model that dramatically improves underwriting quality.

This article was published July 25th 2014 but recently updated September 02, 2014