Predict Customer Behavior With a Virtual Analyst

Marketing would be so much easier if you could somehow predict how visitors will behave through multiple channels. Analytics provide a bit of help, but uncovering the hidden patterns that provide the answers you really need is a major problem.

At the Adobe Digital Marketing Summit in Salt Lake City this week, predictive marketing is a hot topic. I spoke to Matt Langie, director of product marketing for Adobe, who explained why it’s so difficult for marketers to predict customer behavior using analytics alone. He said that with “big data, the massive volumes of analytic data integrated marketers collect through social, through the web, through mobile, video, etc. is so overwhelming that marketers don’t glean a good understanding of this data…At best, analytics is like looking in the rear view mirror. It tells you what happened in your past history, but it doesn’t really do a good job of telling you what’s going to happen in your future.”

So this week, Adobe is announcing new predictive marketing capabilities called Virtual Analyst, which is part of the Adobe Digital Marketing Suite. This new capability is designed to help marketers tackle the “big data” issue, including giving them a tool to guide them in forward-looking decision-making.

For instance, this new tool promises to help marketers build out what-if scenarios to measure how much more revenue they can drive, conversion or adoption rates they might see, and more. A predictive marketing dashboard allows marketers to evaluate how changes will potentially impact business outcomes.

Many years ago, before the glut of big data, I worked alongside conversion optimization wizards Bryan and Jeffrey Eisenberg, who successfully crafted a system called Persuasion Architecture. That system, in part, consisted of predictive scenarios. These scenarios were designed to help us understand what subsets of website visitors wanted to do on the site, and allowed us to establish trackable metrics to match actual scenario behaviors. From this analysis, we were able to significantly optimize conversions. It was a very resource-intensive process, but in the early days, this was groundbreaking.

Today, multi-channel marketing requires even more power and certainly the capability to make decisions based on predictive scenarios across all channels. Virtual Analyst intelligently identifies patterns and develops optimized media mix models to help marketers drive face-forward decisions to provide real business impact.