CIO

AI helps Harmoney improve ability to assess credit risk

‘We have decreased the time required to deploy predictive models from 12 to 16 weeks to minutes’



This is just the next step in our process of disrupting the ANZ banking industry

Brad Hagstrom, Harmoney.

 Harmoney is working with DataRobot to improve the performance of its credit risk assessment process.

 “With our deployment of DataRobot, we’re now using artificial intelligence to reduce risk for our lenders,” says Brad Hagstrom, joint-CEO of Harmoney.

“Our marketplace, which has more than 15,000 members and has facilitated more than $700 million in loans, will now feature the same credit risk assessment capabilities used by the best banks in the world”.

 Harmoney is a peer-to-peer lending marketplace, acting as an intermediary between borrowers and lenders. 

DataRobot will provide provide an automated machine learning platform that puts the power of machine learning into the hands of any business user. It automates the data science workflow, enabling users to build and deploy highly accurate predictive models in a fraction of the time of traditional methods.

 “The machine learning models that Harmoney has created with DataRobot are trained on data captured from more than 300,000 loan applications,” says Hagstrom.

“These models have proven to be so accurate in their real-time predictions of credit default that Harmoney has been able to improve profitability for lenders, reduce costs to borrowers, and sharpen the company's competitive position against incumbent lenders in our market.”

 “With DataRobot, we have decreased the time required to deploy predictive models from 12 to 16 weeks to minutes,” he adds.

 Brad Hagstrom of Harmoney
Brad Hagstrom of Harmoney

The machine learning models are trained on data captured from more than 300,000 loan applications

Jeremy Achin, CEO of DataRobot, says AI will help Harmoney keep prices low for borrowers and returns high for lenders. A lending marketplace should have it no other way.”

 In the three years since its inception, Harmoney has introduced the concepts of a shared economy to New Zealand and Australia’s financial market.

With the application of machine learning automation and the most advanced modeling capabilities in existence, Harmoney is strengthening the marketplace with industry-leading credit risk assessment, says Hagstrom.

““This is just the next step in our process of disrupting both the New Zealand and Australian banking industry and the behemoth financial institutions that currently dominate it,” he states.

“We launched into the Australian market in March 2017 with A$200 million to lend with a borrower-only offer focusing on great rates and customer service for unsecured loans between $5,000 and $35,000 on 36 and 60 month terms.

One year later, we now now offer rates as low as 6.99 per cent and up to $70,000 in unsecured loans on three to five year terms, he says.

“Our Australian business is on a bigger trajectory than the New Zealand business, and we are very excited about the opportunities for lending in both countries with machine learning a contributing factor.”

Send news tips and comments to divina_paredes@idg.co.nz

Follow Divina Paredes on Twitter: @divinap

Follow CIO New Zealand on Twitter:@cio_nz

Join us on Facebook

Sign up for CIO newsletters for regular updates on CIO news, views and events.

Join us on Facebook.