By: Vallerie Miller
In the past, most retailers could only dream of being able to understand what a customer really thought. But that is now not only possible, but also on display in the form of business intelligence tools from Microsoft at this week’s Retail Realm 2016 in Las Vegas.
The conference offered several showcases of services from Microsoft’s Cortana Intelligence Suite that allow users to discover insights in their data. The Cortana Intelligence services can be found on the Microsoft Azure portal.
“People are already [signing up and] using them,” says Luis Cabrera, principal program manager of Cortana Intelligence Suite. “Azure Machine Learning Studio and cognitive services are part of the Cortana Intelligence suite. The goal is to find insights in your data, and to find value in your data. So then you can use it to take action. ”
Training the predictive models can give users a valuable way to analyze past data to determine future patterns and probabilities. For example, in medicine, such models can be used to look at the likelihood of a person having – or getting — a disease such as breast cancer. In retail, focused data can be used, via cognitive services, for scenarios as different as recommending products or helping to predict the probability of fraud taking place.
Much of the talk this week was on the importance of data, without which analysis is virtually worthless, Cabrera maintains.
“People say ‘data is gold,’ but it is really like coal. You have to work to make it valuable,” he says.
Until the recent past, the level of data analytics that Microsoft is showing off this week was pretty much the domain of large companies who could employ data scientists to gather, input, and analyze massive amounts of data.
Cortana Intelligence levels the playing field, allowing smaller companies the same types of analytics. “Cognitive services is a turnkey, ready-to-be used solution,” Cabrera, says.
Cortana Intelligence is set up so a large degree of scientific knowledge is really needed on the part of the consumer, and thus there’s no need to have the expense of paying for independent experts, he adds.
“You just bring your data and we create the model for you,” Cabrera says. “You don’t need to do the data screening. We have the data screened for you. So, you just use it directly. You don’t need to do data science or train the model.”
Retailers can use cognitive services for predictive models that accomplish many tasks. For example, making a new product offering to an existing customer. These “recommendations” are generated for the retailer, based on a customer’s past buying habits.
In making the recommendations for the user, the cognitive services provides the application program interface, or API. That also saves a lot of time and money for the Cortana Intelligence customers, Cabrera says. Examples include a text algorithm API, face recognition API, and a speech API. “And there are many more. We have another one called ‘emotion API’,” he says.
The emotions analytics API aims to detect the emotions on a person’s face, while the text analytics looks at key words in the text to determine the author’s mood and true feeling. These are valuable ways for retailers to learn to true sentiment of consumers, Cabrera adds.
“You just bring your transactional data, and based on that, the system can create an algorithm,” he says. “And based on that, the machine can learn by itself.”
At Retail Realm 2016 at the Four Seasons Hotel, Microsoft discussed integrating Cortana Intelligence into AX 2012 and the new AX (AX 7), with more capabilities likely available in the fall. Cabrera says those moves will make using the cognitive services offering even easier for users.
“The beauty with AX is that the data is already there,” he explains. “So, (we) can extract the data and construct the model without you having to work on all the coding and integration.”
So far, the Microsoft innovation plan is on track to meet its goal, Cabrera notes.
“Our mission is to democratize and advance analytics and make it accessible to not only data scientists who have a lot of training,” he says. “But [we also want] to make it accessible to the people who need it.”