I am a huge proponent of advanced analytics and predictive modeling to improve marketing results.
Moreover, I embrace automated modeling, the integration of clusters (or segments) and modeling, and the use of omnichannel campaign management tools. In most cases, more automation and data and analytics mean better marketing results.
Time after time, we see marketing results soar when consumer data and behavioral data are used to develop predictive models that allow for perfectly timed and customized messages. From health insurance, to life insurance, to retail and e-commerce, to utilities, and even CPG, our advanced analytical solutions can make heroes out of our marketing clients.
To be clear, the power of advanced marketing analytics never ceases to amaze me.
With that said, the advent of machine learning and unsupervised marketing execution has ushered in an era of danger and pitfalls.
Let me be blunt: the use of automated modeling and marketing is not an excuse to turn off your brain.
A Cautionary Tale
Recently, Microsoft's runaway Twitter Bot, called "Tay" serves as a cautionary tale for marketing and analytical professionals everywhere. This article further highlights the need for human participation in the development and execution of marketing models. Both articles are recommended reading for today's direct and digital marketing professionals.
As we all continue to push the boundaries of consumer data, modeling and automated marketing, let's take great care to acknowledge that these times require more from our experience and intellect, not less.