Predictive analytics is one of the most exciting emerging areas in marketing, and technology at large. While there are many wonderful examples below on how brands like Google, Uber and Apple are utilizing predictive analytics, there are far more easy-to-implement and practical use cases. Sailthru by Marigold by Sailthru’s Prediction Manager helps publishers and retail/ecommerce businesses better engage their end-customers and ultimately drive higher engagement and revenue, while optimizing their customer experience at the same time by predicting a customer’s next action, and serving (or suppressing) marketing communications tailored to that customer’s next action. Read more about Prediction Manager here. Want a demo? Let’s get chatting.
—
Smartphones have made it possible for businesses to monitor you at all times. Take a company like Google, for example. You may look up the name of a restaurant over Google Search, make your way there using Google Maps, and perhaps even check the weather and traffic updates along the way. The amount of information you provide to Google here is pretty exhaustive—and a treasure trove in the hands of a data analyst. While privacy advocates are concerned about users providing so much information to commercial entities (like Google), this article looks at the different ways tech businesses transform billions of data points into something extremely useful—perhaps revolutionary. We’re talking about predictive data analytics.
Google Play Music
Music apps have been using historical data to recommend new songs and artists to consumers for a long time now. But Google is doing something extraordinary with Play Music. In 2016, the company launched a revamped Play Music that can recommend music more accurately than any other product out there. It uses dozens of data sources, the main one being the music you have listened to before. But Google also uses a host of other factors to influence your music preferences.
For instance, say you listen to different types of music depending on your activity—maybe classical music at work, upbeat songs at the gym, and perhaps romantic songs while traveling. Google’s machine learning algorithm now interpolates your music preference with other factors like location (at work or the gym, for example), weather (raining or sunny) and even other details pulled from your email or calendar to find the perfect playlist recommendation for you.
Uber Restaurant Guide
As a service that, among other things, transports people to and from restaurants, Uber has pretty valuable data points that can tell which restaurants its customers prefer to visit in any given location. Now, Uber has compiled a restaurant guide based on its own data and other real-time information, such as the number of drop-offs, the types of vehicles used and trending locations. Drop-offs, for example, could give an idea about restaurant popularity and waiting times; the type of vehicle might indicate how upscale the restaurant is; and trending locations could be used to recommend restaurants to customers who don’t have a specific destination in mind. As of now, the Uber restaurant guide only covers twelve cities across the US, although it’ll likely increase in future.
Apple’s Siri Experiment
If there is one product that has brought machine learning to the mainstream, it is probably Siri. The iPhone voice assistant makes use of deep learning (which is a tad different from traditional machine learning) for speech recognition, natural language understanding, execution and voice response. The Siri software has undergone a sea change since it was first incorporated into the iPhone; it now uses machine learning incorporated through deep neural networks, convolutional neural networks, long short-term memory units, gated recurrent units and n-grams to cut down its error rate by a factor of two. Besides Siri itself, Apple also has ingrained machine learning into all of its products, from showing reminders for appointments you forgot to enter on your calendar, to showing hotel map locations before you even type them in. They can even detect fraud on the Apple Store.
Facebook FBLearner Flow
Imagine the amount of data stored and processed on Facebook: it’s humongous. Consider that the earliest users of Facebook now have more than a decade of photos and videos stored on their timelines, which must be pulled up anytime it is requested. Now take into account the billion-plus monthly active users, and the sheer scale of the challenge becomes apparent. In 2015, the company made its AI backbone, FBLearner Flow, available company-wide. This platform is what controls every minute aspect of machine learning and AI within Facebook’s many products. Aside from plainly obvious features like deciding the best content and friends to show on timelines, FBLearner Flow also includes models for many intricate machine learning programs. For instance, one model helps Facebook provide auto-captioning of videos to its advertisers; studies have shown captioned videos bring about higher engagement levels than regular videos and can boost viewing time by as much as 40%. Quite evidently, such machine learning scripts are critical in bringing more advertising revenue. These internal machine learning models have also helped Facebook reduce its reliance on third-party translation tools for its nearly two billion newsfeed items each day (Facebook used Microsoft Bing’s Translation tools previously).
While many machine learning innovations are not immediately evident to layman users (other than as a step towards better user experience), in each of these instances, the companies are facing millions, if not billions, of data points to analyze, execute, test and relearn concepts. It will be interesting to see where these various experiments lead us over the next decade and beyond.
This article is by Aditya Rana from dataconomy.com.