When you think of someone in predictive analytics, which person comes to mind?
Think about it for a moment and then hold that thought…
For me, it most definitely is Eric Siegel. If you don’t already know, Eric Siegel, Ph.D., is the founder of the Predictive Analytics World conference series, the Executive Editor of Predictive Analytics Times, and author of Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die (Revised and Updated).
So when I got the opportunity to pick someone’s brain for a new interview series we’re doing, who else but Eric? Here’s what he had to share about the state of predictive analytics today.
What value does predictive analytics deliver to companies in their digital marketing efforts?
ERIC: Predictive analytics (aka machine learning) targets content, product recommendations, fraud detection, and retention efforts — in all cases, rendering these processes more effective. To get an idea of your possible upside, start with others’ case studies and then do a scratch calculation to forecast your own win. For the first of these two steps, the central insert of my book “Predictive Analytics” is a compendium of 182 mini-case studies divided into nine industry groups, including examples from BBC, Citibank, ConEd, Facebook, Ford, Google, the IRS, Match.com, MTV, PayPal, Pfizer, Spotify, Uber, UPS, Wikipedia, and more.
Where are we in the adoption curve of predictive analytics, as it relates to digital marketing?
ERIC: We don’t have the complete industry data to properly answer that question, but I would informally estimate that we’re about 5-10% where it could and eventually will be, as far as adoption, implementation, and effective deployment. Despite my low estimate, I would say the deployment of machine learning is well beyond early “Innovator” or “Early Adopter” stages. The concepts and technology/solutions are fully developed and proven. But the process to commercially integrate and deploy is not just a technical one – it is an organizational process. This is quite different from most technologies. You need to not only crunch data and derive predictive scores per individual, you then need to actually change the preexisting operational process to make use of the predictive scores, thus fundamentally changing “business as usual”.
What do you see as the biggest challenges in adopting predictive analytics?
ERIC: The greatest pitfall is an organizational/process one. The deployment of predictive analytics is not turnkey or plug-and-play. You don’t just “install” it. Rather, it is a change to organizational processes, priorities, and basic system operations. The per-individual predictions generated by this technology – such as whether an individual will click, buy, defect, commit fraud, or unsubscribe from an email list – are only valuable when acted upon (i.e., integrated into existing systems, thus actively changing “business as usual”). To that end, the project must be conceived up to and including the executive level, and there must be broad organizational buy-in, commitment, and coordination.
What are you most excited about when it comes to the future of predictive analytics?
ERIC: While core technology and software solutions are evolving in exciting ways, I’m most excited about the breadth of business applications, both across digital marketing and beyond (sectors such as financial credit risk and healthcare deploy the same core analytical technology in analogous ways). As the awareness, understanding, and comfort with deploying predictive models grow, so does its organic integration into more and more processes.
Do you know someone else working in predictive analytics? They could be featured in a future post. Tweet at me, connect with me, or email me to let me know.
Predictive Analytics World is coming to New York, London, and Berlin this fall. Don’t miss out!
You can also find Eric Siegel on Twitter and LinkedIn.
Clickstream Evangelist at Syntasa (an AI assisted customer intelligence platform)
Kerry Hew loves sharing ideas with other curious and growth-minded Digital Marketers and Data Scientists alike, and working together to execute on their visions faster.