Skip to Content
Author's profile photo Former Member

Toward a methodology for measuring analytic success

Fair Isaac Corporation is promoting a methodology called “Decision Yield” for measuring analytic success, specifically in-line (aka operational) analytics.  There doesn’t seem to be a lot of meat there (the Harvard Business Review article cited turns out to be written by another Fair Isaac employee, and probably shorter than the blog post itself I linked to above), but it’s worth a look just for consciousness-raising.  Highlights include:

Decision Yield’s holistic approach involves comparing five different dimensions of decision effectiveness, By considering all these aspects the Decision Yield approach allows you to make a comprehensive assessment of an operational decision. The five areas are:      * Precision or how optimal is the decision?     * Consistency or how consistent across divisions and channels and time is the decision?     * Agility or how quickly can you effectively change the decision when you need to?     * Speed or how quickly can you make the decision?     * Cost or how much does it cost you to take the decision?

While Decision Yield is a fairly general-purpose tool, it is designed specifically to evaluate automated decisions that are typically:      * Customer-facing—from approving loans to pricing insurance to determining cross-sell offers.     * Very frequent—often many thousands of times a day.     * Executed in real-time (credit over-limit approval) or in batch mode (matching an offer with a prospect).     * Delivered through another system such as a website, call center system or letter shop. Often the same decision is delivered through many such systems.

Assigned Tags

      Be the first to leave a comment
      You must be Logged on to comment or reply to a post.