CredoLab

Partner

CredoLab develops bank-grade digital scorecards built on mobile device data. Their pay-per-use solutions are available to banks, consumer finance companies, auto lenders, online and mobile lenders, insurance companies, and retailers. Established in 2016 in Singapore, CredoLab has 46 customers and is operational in 12 countries.

Region

APAC

Specialties

Credit scoring & Decisioning

CredoLab and Mambu use smartphone device data to generate a behavioural score for the customer that predicts her probability of defaulting. By accessing anonymised, metadata of the device, our AI-powered algorithm sends the financial institution the score of the applicant in a matter of seconds. In doing so, not only we open a whole new segment of customers for banks and lenders, but we also bring down their cost of risk, speed up the processing time, and increase their overall predictive power (i.e.: higher Gini Coefficient).

Core to CredoLab’s algorithm is its modelling pipeline. The data processing pipeline consists of a series of automated steps, rooted in machine learning techniques, that ultimately outputs a predictive model for credit default. To protect the confidentiality and to ensure against bias towards individual loan customers, only non-identifying metadata is used.

CredoLab offers a set of easy to integrate, plug-and-play solutions that give decision makers all the information they need in real-time. The data collection is done through either a standalone mobile app, the CredoApp, or through a light CredoSDK library that integrates seamlessly with the financial institution’s existing app. Once installed, and the explicit data privacy consent is given by the customer, the app proceeds to analyse the anonymous data.

This offering is the only provider of bank-grade credit scoring solutions based on smartphone device data. None of the applicant’s personal information nor social media data is used. The algorithm reads and analyses the device metadata without moving any files or information out of the phone. The result of the analysis is sent exclusively to the financial institution. Nothing is sold or shared with any third party, and no ads are ever displayed.

The CredoScore algorithm churns all the data fed into it from the device into a score that can be applied to banks, automobiles, insurance or any other business to improve the quality of decisions taken.

Key benefits:

  • Super light IT integration required.
  • Lower cost of risk with tailor-made digital scorecards.
  • A faster decision with a 1-minute lightning fast time-to-yes.
  • Use behavioural data to assess customers’ willingness to repay and enable risk-based pricing.
  • Increase market share in target segments not covered by traditional credit bureaus.

What sets CredoLab apart:

  • Data sources: CredoLab uses non-intrusive metadata to arrive at the credit score. No private or sensitive information of the applicant is accessed or moved out of the mobile phone.
  • One-time analysis: The algorithm analyses metadata only once, at the time of application, not perpetually. Metadata is accessed only after the explicit privacy consent is received from the applicant.
  • Scalability: The AI-based algorithm learns and becomes more accurate and predictable as more data is fed into it. Therefore, the larger the set of data observed, the better the scoring. As the financial institution grows market share and number of customers, CredoLab calibrates the model to ensure high reliability and stability.
  • Privacy: Applicant’s metadata is of the utmost importance. CredoLab explains what metadata is accessed, why, and how it gets processed. The Privacy Policy is publicly accessible to all users in multiple languages. Furthermore, CredoLab’s app does not proceed with any credit analysis unless the privacy consent and the necessary mobile operating system’s permissions are given. Once these are received, the app captures only metadata, not personal data. Once the metadata is analysed, the results is shared only with the lender tagged to a unique, randomly generated identifier, and not shared with any third party.

For more information, visit partner website.

shape
shape
shape
shape
shape
shape