Why It’s Dangerous to Leave Business Experts Out of Predictive Analytics Projects

Loraine Lawson
Slide Show

Six Big Business Intelligence Mistakes

Both the business and IT sides may be getting ahead of themselves when it comes to advanced and predictive analytics. Maybe it’s time to slow down and remember that what makes technology projects successful isn’t the technology, but how well the business and IT can align on implementing it.

“We now face a new organizational Wild West, brought on by the promise of Big Data and increasingly sophisticated predictive analytics,” A recent CIO.com article opined. “Multiple uncoordinated experiments and campaigns can yield impressive learning, agility. But they can also lead to unintended consequences.”

The recommended solution for taming this beast is nothing flashy or new. It’s simple, old-fashioned alignment between IT and business. This has been a problem for both sides, actually. Business divisions are launching their own analytics in the cloud programs, and as I’ve argued previously, the odds are good that they aren’t consulting IT first. This approach has always lead to (and will no doubt continue to lead to) data silos and integration problems down the road.


For it’s part, IT needs to target its analytics work to what matters to executive leadership. But the alignment shouldn’t end there. Particularly with predictive analytics, the goal is to use the data to change business outcomes or trigger a reaction. It’s critical that IT work with business leaders to determine what those outcomes and alerts should be, rather than assuming they know what will work.

In my previous post, I wrote about a successful predictive analytics project identifying which early indicators determine whether a student will drop a college course. The project was lead by Bill Thirsk, vice president of IT/CIO at Marist College, in New York. It’s been so successful that other colleges are using the model.

The project had a moment, though, where it could have failed because IT acted on its own rather than consulting teachers about how to put students back on track. Once IT found that the model could successfully identify at-risk students, the developers set the system up to trigger emails to the students advising them that they were at risk of failing the class. The result surprised IT: Students dropped out at a higher rate.

Data Analytics

"That wasn't the intent of the research,” Thirsk said. “The intent of the research was not only to try to figure out how to not only alert the student but help them persist to success.”

The division then turned to Marist’s School of Education for advice and, not surprisingly, they were “completely appalled,” he said.

“They came back and said, ‘Look there's more than just a student not paying attention or not being engaged. There might be other things that are going on in their life.’”

Now, instead of messaging the student, the system sends an alert to the faculty, who can then sit down with the student to learn more. As a result, the numbers have flipped, with 60 percent of students sticking with the class to completion.

Loraine Lawson is a veteran technology reporter and blogger. She currently writes the Integration blog for IT Business Edge, which covers all aspects of integration technology, including data governance and best practices. She has also covered IT/Business Alignment and IT Security for IT Business Edge. Before becoming a freelance writer, Lawson worked at TechRepublic as a site editor and writer, covering mobile, IT management, IT security and other technology trends. Previously, she was a webmaster at the Kentucky Transportation Cabinet and a newspaper journalist. Follow Lawson at Google+ and on Twitter.



Add Comment      Leave a comment on this blog post

Post a comment

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

null
null

 

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.