“Gone are the days where technical people only work in IT,” Patrick O’Hagan, director of market planning at Starbucks, told the Level cohort in Seattle this past Tuesday. With that O’Hagan kicked off a lecture with his colleague Laurence Norton – Director of Business Intelligence Strategy and Solutions – about where business and data analytics meet in the real world.

With the digital universe growing exponentially – doubling in size every year according to a recent EMC report – companies are increasingly creating data at every stage of their operations. Few however have built the capacities to analyze and utilize this data. According to the same report, 37% of data created would be useful if analyzed but a meagre 1% is actually analyzed.

O’Hagan and Norton provided insight into how Starbucks is using the power of data to drive its decision-making process. Through a system called Atlas, Starbucks links to as many external and internal APIs as possible, connecting the data with R to build cannibalization models that can determine impact to existing stores if a new store enters the area. This model drives decision-making in cities across the US and world.

Starbucks is particularly interested in determining what impact weather has on sales. To answer this Starbucks combines APIs with information on weather reports and sales analyze the results and create a model to map the impact. Now any internal Starbucks user who wants can see the impact of weather on sales can according to O’Hagan.

In the latter part of their talk O’Hagan and Norton talked about the importance of telling a story with your data with Norton describing how Starbucks’ “models have to be seamless, acknowledging that the end user may not be a very technical person.” O’Hagan chimed in, agreeing and adding that “the hardest thing for people is to talk about data in a human way, so non-statisticians can understand. This is especially important at Starbucks.”

In the final part of their talk O’Hagan and Norton described how Starbucks is making the transition from a retail company to a data analytics one, and what effect this is having on their hiring process. O’Hagan described how he is interested in the “new school” of data analysts and scientists who use R, Python and Tableau among other tools to do their analysis. Norton added some details from his personal experience at Starbucks – where he has worked for a decade – explaining how he has seen a shift from hires primarily work with SQL to hires who are fluent with a wide variety of tools.

Overall, O’Hagan and Norton reaffirmed the trend towards increased integration of data at all levels of decision-making in companies.