Evergage Announces Decisions and Data Science Workbench
Evergage today announced Decisions and Data Science Workbench, two new offerings powered by machine learning and aimed at empowering companies to offer more personalized experiences to customers, as well as analyze how effective those efforts have been.
Decisions leverages artificial intelligence to automatically deliver the most relevant content or experience to a website visitor, app user, or email recipient. It will eventually be a suite of algorithms, with the first algorithm—Contextual Bandit—announced today. Contextual Bandit does two key things. First, it estimates the probability of a person interacting with each available offer or experience on a given channel in real time. Second, it uses machine learning to predict the content with the highest value return by weighing the probability of a person accepting a particular offer or promotion with the business value of that offer to the company.
“Decisions is an advancement in our machine learning capabilities in terms of delivering offers, promotions, and images based on what we know about an individual while also taking into consideration what the best offer or option would be for the company,” says T.J. Prebil, director of product marketing at Evergage.
“If you think about most companies’ websites and their home pages, generally they have a number of audiences that they’re trying to connect with, and then images or offers that support each of those audiences. Trying to define which offer to show somebody can be a challenge,” Prebil says. “Contextual Bandit is designed to solve for that problem where you can upload or associate a series of images in an algorithm and then set that for a particular area of your website, and it will use what it knows about the individual and the dollar amount associated with a particular image to figure out which image is the best one to show someone.”
Data Science Workbench makes the data stored on the Evergage servers available to data scientists in an environment that supports activities such as data transformation, numerical simulation, statistical modeling, and data visualization. The data—which is collected in a unified profile for each customer, visitor, and account that a company has—includes behavioral data, contextual information, and survey responses, as well as first- and third-party information from other systems such as CRMs, email service providers, and data warehouses. Data Science Workbench is a component of the Evergage Gears framework, which was announced in September 2018 and allows companies to extend the core capabilities of the Evergage platform.
“The concept behind Data Science Workbench is that Evergage attracts very detailed and in-depth customer data as people are interacting with a website or a mobile app, as well as transactional information and product catalog engagement,” Prebil says. “That’s all really valuable information for companies and we use it to run personalization, but for data scientists who want to get access to that stuff, they haven’t had the ability to do so. What we’re doing is opening up Evergage to data scientists so they can now have access to our rich customer data that they can pull into an environment that they’re familiar with…so they can their own analyses [on it].”