Helping Our Customers Better Understand Credit

January 17 2017 The Lendable Blog

Data analytics and machine learning have always been at the very core of what Lendable does—from developing our proprietary Risk Engine to building software to support our standard receivable financing structure. Up until this point, however, Lendable has focused on building tools to help investors and originators transact. While our Risk Engine is capable of much more, we hadn’t built the interfaces for users to leverage it. Now, to provide more value to our originator customers, we’re launching Lendable Labs.

Lendable Labs is an initiative to build apps and services on top of our industry-leading data science platform, providing tools and best practices for originators to get better at what they do. Our aim is to help credit-providing firms run their businesses more efficiently, whether they are a PAYG solar provider who wants to improve credit quality or an MFI looking to improve back-office operations.

Ok, we are being a little selfish as well. Not only will Lendable labs enable us to offer investors better investment opportunities, but it will make us a more appealing option for our originators. All in all, it seemed like a no brainer.

Lendable Labs’ first product: Customer Quality Score (or CQS)

Most originators don’t have enough credit officers to follow up on every late customer. Knowing where to place limited credit resources is hard and there aren’t tools to help. Lendable Labs’ first app is a tool to help prioritize collection activities by understanding which customers are most likely to fall behind. With this tool, originators will be better able to manage the ever-present dilemma: when customers are behind on their loans, who do you contact first?

While most companies have some way to prioritize collections, their approach is typically not sophisticated: ranking customers based on how much they owe and how late they are on their payments. Lendable’s tool creates a probabilistic score to add accuracy and provide simplicity. This score predicts the likelihood that a customer will pay, when, and how much. For originators, this can mean a more efficient and automated workflow in their collections offices, adding tens of hours back into the week for their staff.

We are in the process of piloting our CQS tool with two originators, one in East Africa and one in West Africa, capturing insights and evidence on the increase in rate of repayment using this tool. We will also be publishing white papers on the results of these pilots and designing the final version of the tool based on these learnings.

Our goal with Lendable Labs is to offer valuable analytical resources to our partners and alternative lenders around the world. If we can help lenders make credit and collections decisions more effectively, then it is a success.

Interested in learning more about Lendable Labs or partnering with us? Get in touch at

A special thanks to The David and Lucile Packard Foundation for supporting the creation of Lendable Labs.