The Data Analytics Blog

Our news and views relating to Data Analytics, Big Data, Machine Learning, and the world of Credit.

All Posts

Frequently Asked Questions On Credit Scoring And DecisionSmart

March 13, 2019 at 7:56 AM

We’ve written previously on how Business Rules Management Systems (BRMS) like DecisionSmart are used across the customer lifecycle, and implementing scorecards is a function used in multiple lifecycle stages.

In this blog, we discuss some of the most common FAQs when implementing credit scoring in our proprietary BRMS, DecisionSmart.

What is DecisionSmart?

DecisionSmart is a decision engine that allows business users to capture and apply business rules, matrices, scorecards, decision trees, terms of business tables and decision flows across the entire customer lifecycle.

Embed video

Who develops the credit scorecards that are deployed in DecisionSmart?

Principa develops all scorecards in-house. We are able to provide various options based on the availability as well as volume of historical application data available.

Is the model user interface self-explanatory and user-friendly?

DecisionSmart is fully business configurable. DS-1A screenshot of the scorecard metaphor is shown alongside. Characteristics can be added from the data dictionary; attribute groups can be defined; and weights input/adjusted.  For scorecard alignment parameterised values are given too, allowing for effective scorecard management.

Can DecisionSmart be customised to any local language requirements other than English?

DecisionSmart modules are not utilised by an operations team, the configuration is done by business in English on these systems. However, there are certain fields and components that can be translated into different languages to ensure when the information is passed back into the Loan Origination System, it is understood by the user.

Is DecisionSmart open to APIs?

Yes, the API used is SOAP.

Under what circumstances would you consider using a hard rule over a scorecard?

There are many circumstances where “hard” policy rules would be suitable to be utilised over a scorecard. Some of them include:

  • Legal rules (e.g. under-age, unemployed, on anti-money laundering list, on terrorism list etc.)
  • Reckless lending rules – if a legal requirement or bank requirement (e.g. 2 current judgments on the credit bureau, currently 2 active accounts 3+ in arrears)
  • Affordability rules – also reckless lending (e.g. applicant’s affordability too low)
  • Internal bank data – previous fraud, previously written-off by your company, currently in-arrears on another account at the bank
  • Verification rules – e.g. identification document and given ID number not the same

How often are scorecards recalibrated to optimise performance?

A new expert scorecard should be assessed after a maximum of six months. If recalibration is required, then that can take place there on “early-performance”. After twelve months a full calibration on full performance is possible. Thereafter scorecards should be monitored ideally every 6 months (maximum 12 months).  Some scorecards remain stable for some time due to a stable and predictable population; others deteriorate quickly when a population changes. Scorecard recalibration or redevelopment is very much dependent on this and it may differ across products and across countries.

To determine a client’s affordability, what sources of data would you require?

This depends on the specific rules within the countries.  Affordability normally comprises the following:

  • Income of applicant (or household income)
  • Debt commitments (normally ratified through a credit bureau, although this is difficult in countries with limited data)
  • Household expenses (normally declared by the customer or subject to a formula taking into account demographic data and income)
  • Proof of income – via pay-slips or bank statements

What operating systems does your solution utilise for the various components and what version of these operating systems is recommended?

The SmartSuite will need to be deployed on a IIS server supporting MS SQL 2016. The application and database server recommended specs are: 2GHz 64-bit processor, 32GB memory with 200GB of disk space (500GB for database server).
Contact Us to Discuss Your Business Requirements

Latest Posts

Frequently Asked Questions On Credit Scoring And DecisionSmart

We’ve written previously on how Business Rules Management Systems (BRMS) like DecisionSmart are used across the customer lifecycle, and implementing scorecards is a function used in multiple lifecycle stages.

Top 5 LinkedIn Groups To Join On Data Analytics

With LinkedIn usage growing by two new members every second, you simply can’t afford to not be on the platform. Founded in 2003, LinkedIn has 590 million users with 260 million of those active every month.

The EQ Behind The IQ: The Growing Disconnect Between The Data-Driven Business And It's Customers

The EQ Behind the IQ - Jaco Rossouw from PrincipaDecisions