Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
Survey: Machine Learning Projects Still Routinely Fail to Deploy
 Originally published in KDnuggets. Eric Siegel highlights the chronic...
Three Best Practices for Unilever’s Global Analytics Initiatives
    This article from Morgan Vawter, Global Vice...
Getting Machine Learning Projects from Idea to Execution
 Originally published in Harvard Business Review Machine learning might...
Eric Siegel on Bloomberg Businessweek
  Listen to Eric Siegel, former Columbia University Professor,...
SHARE THIS:

9 years ago
Good Predictions != Good Decisions

 

A Fateful Tale

Ted is having a rough week at work. As a call-center employee, his main focus is on customer retention and offering promotions to his company’s current subscribers. For some reason, his numbers are horrible this week. Not only are customers rejecting his promotions, they’re actually canceling their subscriptions entirely! It’s only Wednesday, and he’s had as many customers leave in the past three days as in the previous month. He’s nervous as he enters the conference room with his coworkers for their mid-week meeting.

Ted’s nerves start to subside as he realizes that others are in the same predicament. Customers are churning left and right, and the manager is visibly upset. As it turns out, their abysmal numbers are the result of a change to their weekly call lists based on a mandate from the corporate office; some analysts had built a model that was supposed to provide the “best” customers to target with their phone calls. These analysts were obviously wrong. “How can they know how customers are going to react when they’ve never even talked to one?” Ted murmurs.

This content is restricted to site members. If you are an existing user, please log in on the right (desktop) or below (mobile). If not, register today and gain free access to original content and industry news. See the details here.

Comments are closed.