Adobe’s new Virtual Analyst doesn’t need questions to provide answers
The AI-powered addition to the Analytics Cloud is designed to alert marketers to data insights without being asked.
The best helper is the one who knows what you’re looking for before you ask.
That’s the idea behind Adobe’s new Virtual Analyst in its Analytics Cloud, announced Monday. It employs the company’s Sensei artificial intelligence layer to find the “unknown unknowns,” even when the user hasn’t asked for that specific info.
How it works. For instance, the company says, the Analyst might point out which audience segments in which locations at which time are buying more of specific product models, or reasons why there is an unexpected drop in online orders.
With “like” and “don’t like” feedback for the recommendations, the Analyst learns over time what kind of insights are more valuable to a marketer.
John Bates, director of product management at Adobe Analytics, said that only about 1 to 3 percent of all the data available to marketers is generally unlocked for insights, because of the difficulty of finding the right questions to ask.
The vision for a data helper. The Virtual Analyst is Adobe’s latest step in its effort to sift the mountains of data that marketers acquire. In 2013, Analytics Cloud released Anomaly Detection, which finds “statistically significant” deviations of many kinds. And alerts have previously been available, but the marketer has needed to manually create them.
The vision for the Virtual Analyst, Bates said, is to “go beyond questions I asked to questions I didn’t ask,” automatically creating alerts and making them available.
Opinions expressed in this article are those of the guest author and not necessarily MarTech. Staff authors are listed here.