Christopher S. Penn is chief innovator of Brain+Trust Insights, where he's responsible for the creation of products and services, technology and marketing strategy, brand awareness, and research & development. He's the author of Marketing Blue Belt and Leading Innovation, and co-host of the Marketing Over Coffee podcast with John J. Wall. Chris is also a well known Google Analytics expert (check out his course, Marketing Plan Framework: How to Build a Data-Driven Customer Journey).

Back in 2016 on Marketing Smarts, Chris talked with me about innovation in the workplace. This time around, I invited Chris to discuss predictive analytics: what "predictive" means in the analytics context, how predictive differs from "prescriptive" analytics, and what you can do to take advantage of the additional insight predictive offers.

Here are just a few highlights from our conversation:

"Predictive analytics" might sound intimidating, but it's just using data to plan your future marketing (02:54): "Predictive analytics is exactly what it sounds like: using analytics to predict things, which is really important for us as marketers. We have a tendency in marketing to drive with the rear-view mirror, which is difficult to do, but that's really what you're doing when you're using analytics and only looking backwards at what happened. The ability to use predictive analytics helps us to plan ahead, not for the unforeseen but for the foreseen. It takes some of the uncertainty out."

Predictive analytics helps you to foresee what will happen; prescriptive analytics suggest what you ought to do about it (07:30): "The difference between predictive and prescriptive is [with predictive] you know what's going to happen but you haven't decided what to do about it. With prescriptive analytics, you should be able to do some mathematics to figure out what should we do next, what does the data tell us. The machines can do that very tactically, but this is really where humans...still have a future and jobs, at least of the near term, because a lot of that requires judgment. A lot of that requires experience across multiple domains.

"A machine might see, for example, a sharp dip in the prediction for December 25th and might recommend 'up your bid prices 400% that day to compensate,' and we all know people are opening presents and drinking eggnog. No amount of bidding up is going to fix the fact that there's a cultural holiday and the machines are not aware of that. That is where prescriptive analytics is human-run, informed by the data that the machines put together."

Most marketers are not ready to take advantage of predictive analytics using current technology (08:41): "[Marketers] are not ready at all across the board [for prescriptive analytics]. My gut feeling is that they probably never will be, in the sense that I don't see a ton of marketers running out to learn how to code, which is what you need to do to build a lot of stuff.