Policy First Attribution
Attributes are tough to add to the inner workings of policy execution technology. That’s where most data acquisition efforts fall off the rails. Those in charge of the data can say “we have the data” and the decisioning system can say “it’s just not affecting much.” Suddenly the 10 basis point lift or five percent acceptance jump you were sold by your data vendor is cut by 75 percent.
Effective attribution requires a “business first” attribution implementation framework. If your attribution is in C code (or perhaps even COBOL) you are doing it wrong. If it can’t be read by a statistician, you are doing it wrong. If it doesn’t spit out validations at the push of a button, you are doing it wrong. Best practice systems are easily auditable. One glitch in implementation and the policy won’t work. Lack of access to unit-testable test cases means working and reworking the policy. It is easy to call these sorts of items the “minutia” of a system, but when looking for the culprit in why a policy took nine months, instead of one, to implement, these are your common causes.
Adding new data is the future. How it integrates into your system, your policy, and your customer interactions is dependent on the first system in the flow. We recently worked with a telecommunications provider that was trying to add a new telecommunications data source with a custom model. The whole process took all of three months from our end because the telco was already using our data acquisition framework. Others are still trying to figure out the kinks some 9 months after the data source was launched.
In our next blog, we will address improving data acquisition.
If you want to talk to an Equifax specialist about managed services, please send us an e-mail.
If you are interested in a monthly e-mails about blog updates, sign up for our newsletter
This post was contributed by:
Recommended For You
New Podcast Examines Latest Economic Trends for Consumer Credit and Small Business The Market Pulse podcast is a new production from […]
Buy now, pay later FinTechs face merchant risks and consumer risks simultaneously Thank you for joining me for the second […]
Why Proxy Model Explanations of Neural Networks Inevitably Fail For financial service providers, credit risk modeling is all about identifying […]
The Past as a Window to the Future In my last post, we spent time exploring methods to mitigate the […]