What is “Analytics”
Analytics is the process of using data relationships and computer models to drive business value, improve decision-making, and understand human relationships.
If the Information Age began in the 1990s with the rise of technology, then we’ve now officially entered the “data age,” where companies like Google, Facebook, IBM, Teradata, Oracle, and SAS have the capacity to gather a lifetime’s worth of data about their customers and the customers’ behavior.
But all that data is just a massive pile of numbers until a skilled analyst turns those numbers into meaningful information useful for making intelligent business decisions. Today, companies are searing for experts in data analytics with high formed business and technology backgrounds who understand the importance of the latest data and information-age trends.
This requires more than simple data analysis. Prescriptive analytics focuses on trends using simulation and optimization, while predictive analytics uses statistical tools to predict the future, and descriptive analytics is about enabling smart decisions based on data.
Data miners and data analytics experts who are versatile in all three areas of analytics help corporate executives translate their data into intelligent information, which provides companies a competitive advantage and increases their bottom lines.
Prescriptive analytics is about enabling smart decisions based on data
Why do airline prices change every hour?
Basic economics teaches us that higher demand drives higher prices. So, if you knew when the least desirable flying days and times were, you'd easily know when to book the airline tickets for your next vacation in order to get the lowest price. Turns out that the airlines have a leg up on consumers when it comes to this information; they use "prescriptive analytics" to sift through millions and millions of flight itineraries instantaneously. They use this data to set an optimal price at any given time, based on supply and demand, thus maximizing their profits.
Prescriptive analytics helps airlines squeeze every dollar out of passengers' wallets, making sure that the highest prices are charged during the highest times of demand. Airlines even take calculated gambles by deliberately withholding cheap fares during low-demand times in anticipation of a future higher-paying passenger. Analytics is key in helping industries like the airlines ensure that their pricing structures are working hard to contribute to bottom-line results.
Why does Facebook often find your acquaintance as potential friends?
Imagine that every Facebook user is represented by a dot. Now imagine drawing a straight line between every Facebook user and each of his or her friends. With over 750 million Facebook users, you would have quite a chaotic drawing of intersecting lines.
This is where prescriptive analytics comes in - to create order out of the chaos and help Facebook recommend the right friends for you. It works like this: If you and your friend John Doe have many friends in common, then your straight lines and John's lines have common endpoints. But if John has a friend who is not on your list of Facebook friends, it is very likely that you know that person. Prescriptive analytics facilitates scanning of billions of such lines to determine possible missing friendships. So thanks to analytics, we're all able to find our school buddies or long-lost childhood friends.
Predictive analytics is predicting the future based on historical patterns.
How do grocery cashiers know to hand you coupons you might actually use?
Each Tuesday, you head to the grocery store and fill up your cart. The cashier scans your items, then hands you a coupon - for 50cents off your favorite brand of whole-grain cereal, which you didn't get today but were planning to buy next week.
With hundreds of thousands of grocery items on the shelves, how do stores know what you're most likely to buy? Computers using predictive analytics are able to crunch terabytes and terabytes of a consumer's historical purchases to figure out that your favorite whole-grain cereal was the one item missing from your shopping basket that week. Further, the computer matches your past cereal history to ongoing promotions in the store, and bingo - you receive a coupon for the item you are most likely to buy.
Why were the Oakland A's so successful in the early 2000s, despite a low payroll?
During the early 2000s, the New York Yankees were the most acclaimed team in Major League Baseball. But on the other side of the continent, the Oakland A's were racking up success after success during the same time period, with much less fanfare - and with much less money.
While the Yankees paid its star players tens of millions, the A's managed to be successful with a low payroll. How did they do it? When signing players, they didn't just look at basic productivity values such as RBIs, home runs, and earned-run averages. Instead, they analyzed hundreds of detailed statistics from every player and every game, attempting to predict future performance and production. Some statistics were even obtained from video of games by using video recognition techniques. This allowed the team to sign great players who may have been lesser known but were equally productive on the field. The A's started a trend, and predictive analytics began to penetrate the world of sports with a splash, with copycats using similar techniques. Possibly predictive analytics will help bring Major League salaries into line sometime soon?
Descriptive analytics mines data to provide business insights.
How does Netflix frequently recommend just the right movie?
Netflix has tens of millions of users, each with their own movie preferences. Let's say you watched two movies this past weekend, and they were both dramas. Across the Netflix universe, many other people watched similar dramas to the ones you chose. Then, next Saturday, they chose a third movie, which might also have been a drama. Based on this information about user preferences, Netflix predicts that you would likely want to watch a drama that's similar to the third movie chosen by others who have similar taste to yours.
Netflix uses descriptive analytics to find correlations among different movies that subscribers rent. Movies have many attributes, including genre, rating, length, subject matter, and so on. With so many users and so many attributes across the Netflix's spectrum, obtaining a recommendation specifically for you with seconds is a daunting task. But analytics helps confine the universe of movies' attributes to a small number, yet still capture most of the relationships to help build a set of preference data. Descriptive analytics helps companies like Netflix make sense of the millions of choices its users make every day.
Why is it better to charge an electric vehicle overnight and not during the day?
We know that electricity prices are the highest during times of peak energy demand. But when are those times? Intuition might tell us that peak demand often happens in the late afternoon, when everyone comes home from their day and turns on lights, air conditioners, washing machines, televisions, and computers. It might also tell us the lowest demand occurs while we sleep at night.
Descriptive analytics examines historical electricity usage data to confirm our suspicions. This type of data analysis also helps electric companies set prices, which sometimes means that electricity rates during the low-demand nighttime hours might even be negative! What a great time to charge your electric vehicle - you're not only reducing pollution, but you even get paid for using electricity to charge your car. As electric vehicles gain traction in the marketplace, it will be interesting to see if they create a second peak during the night. Data analytics will help us find the answer.