The Goal: Financial institutions are targeting millions of consumers who do not have robust credit histories or have not used credit cards but are potentially great customers. Ensuring that these people and groups have access to a range of quality, affordable, and appropriate financial services is on the agenda of marketing teams and credit risk departments. These teams leverage analytics to generate a more complete understanding of households’ financial needs.
The Requirements: Alternative credit scoring is required to service this target population, since traditional credit scoring does not apply. An array of nontraditional data sets must be gathered from diverse sources and in very large volumes. For example, an organization may want to use data gathered from mobile operators, grocery stores, and utilities to compute predictive analytics that may determine if a borrower will pay for insurance or repay a loan on time.
The Challenge: Many consumer lenders have advanced credit-risk modeling capabilities, but accessing raw sensitive behavioural and transactional data from alternative sources require changes technologies and approach. Customer activity data is sensitive, and each data owner will allow it to be combined and commingled with yours so long as they can control how and by whom the data is used. Analytics require this detail but cannot exposing such sensitive data in violation of the owner’s corporate policies and regulations.
The Problems: Conventional systems can be configured to allow users to expose or change data owned by another party. If data Owner A shares raw data with Owner B, using a conventional system, then Owner B can use the combined data sets in violation of Owner A’s rules. Legal agreements cannot be relied upon to prevent data misuse. Most data sharing projects get stuck as a result. Reliance on 3rd party services do not provide the capability for self-service analytics, and do not provide the data quality and transparency required to ensure the analytics are generating accurate results.
The Myngl Concordance Solution: Secure Analytic Containers work differently than conventional systems which allow users to view or change data from another owner. Concordance decentralizes security, governance and compliance controls to each data owner separately and simultaneously. Users, admins, even Myngl staff, are blocked from direct source data access to change or override owners’ rules. This removes vulnerabilities and prevents data risks, misuse, and breaches.
Analysts can make use of sensitive data without exposing it. Secure Analytic Containers provide analysts with a secure analytic sandbox utilizing open-source Zeppelin Notebooks, SQL, Python, TensorFlow and other powerful tools. Analytic jobs are separately authorized by data owners to make use of restricted data. Automated scheduling is orchestrated via microservices and results are segregated from source data. This allows analytic apps to use data elements which are hidden from direct Subscriber view.