Have you heard of the tech singularity? Chances are you have. But just as a reminder, the singularity is the hypothetical point at which artificial intelligence becomes so intelligent it recreates itself. Then, it becomes so complex we can no longer control, predict, or comprehend it. This point—the unknown known of technology creation—is the singularity.
Have you heard of the tech singularity? Chances are you have. But just as a reminder, the singularity is the hypothetical point at which artificial intelligence becomes so intelligent it recreates itself. Then, it becomes so complex we can no longer control, predict, or comprehend it. This point—the unknown known of technology creation—is the singularity.
Will we ever get there? Is Google’s Deep Mind and AlphaGo—which beat a world champion at the game of Go—an introduction to the singularity? We can only speculate. What we know for sure is that AlphaGo won because of big data. The program’s approach to analyzing and learning from data gave it the edge.
To beat Lee Sedol, AlphaGo used data on previous Go matches he’d played. It then personalized its approach, specific to Sedol’s gameplay. In the same way, those who work with big data are using it zoom ever-closer in on their subjects.
Dynamic personalization
According to Base Creative, an international branding agency, one of 2016’s important marketing trends is dynamic personalization. Personalization delivers five to eight times better Return on Investment (ROI) than non-personalized ads. Data analysis and personalized marketing can boost sales by 10 percent or more. But the challenge is to balance data usage with privacy concerns and “serendipitous discovery”. In other words, marketers can’t forget about the insights they can glean from flesh-and- blood humans.
Yet, data is coming to represent and define the flesh-and-blood human being. Does that sound anything like the beginnings of the singularity to you?
To the consumer, it seems almost absurd that a set of numbers defines them. But to the data scientist, and to the marketer who benefits from that scientist’s work, the accuracy of data is anything but absurd.
Modern Data Science
Any approach to using big data in the marketplace starts with individuals who know how to analyze data. Those individuals must know about data storage and data mining. They’ve got to be able to design customized algorithms and recognize patterns.
For the individual undergoing data science training, the numbers are staggering—but promising for the field: the amount of data triples every year, growing at an annual rate of 40% up until the year 2020.
In the retail sector alone, retailers who use big data stand to improve their operating margins by more than 60%. The data scientist helps the retailer personalize their operation for the people who walk through the door. This involves knowing what data to mine and how to decipher raw data. It involves going further and deeper with analysis until the scientist yields insights about customers. Then, the scientist provides actionable insights.
Modern Marketing with Data
Once the scientist delivers actionable insights, the organization must act on them. For the marketer, increasingly this is this is about personalizing marketing efforts.
According to this infographic from Monetate, 94% of marketers agree online personalization is critical to their business. Companies that use analytics show a 49% increase in revenue growth versus those that don’t. And 35% of marketers say personalization through big data has improved customer engagement.
So what does personalization look like? It’s a targeted approach to marketing—Johnny walks into the mall and sees a promotion pop up on his smartphone about discounts at Pottery Barn. Turns out there’s a Pottery Barn in the mall.
This type of marketing uses geofencing. Since smartphones have GPS installed on them, if the user opts in to certain mobile add preferences, or promotions from a brand, the store can use GPS-triggered alerts to advertise based on location.
You can see a lot of the personalized marketing on social media. According to Simplilearn, Facebook is the number one platform for marketers—52% of social media users prefer it as their social platform.
Facebook collects a great deal of data, and the ways the social network has used it showcase personalization. Facebook’s ‘Flashback’ movies were collections of each user’s moments since joining the network. To compile the videos, Facebook looked at the photos and posts users had shared the most.
During the 2010 midterm elections, Facebook introduced the digital ‘I Voted’ sticker. Users could add it to their profiles, and when other users saw it, 20% of them added it, too. This may have influenced up to 340,000 people to vote who hadn’t already.
According to Facebook’s analytics chief, Ken Rudin, companies need to answer the right questions with data. He says, “A meaningful question is defined as one that leads to an answer that provides a basis for changing behavior.”
Facebook now has topic data, which allows marketers to look at data such audience demographics and sentiment relating to a specific topic. The marketer can ask a meaningful question, such as, ‘what are women ages 21-35 saying about Jose Cuervo tequila?” The marketer can then tailor adds to that age group based on the audience sentiment. This can also drive product development.
Where we’re going
Are we headed to the point where computers will replace data scientists and marketers? Based on previous analysis, will machines know what data to gather and analyze, what patterns to look for, how to customize algorithms, and what insights to deliver?
In fact, that’s what Google’s Deep Mind Artificial Intelligence wants to do for search. It will create algorithms on the fly based on personalization. When you type in a term, it will look at your previous searches, what you’ve clicked on, what others have done, and deliver a customized canon of results.
But machines can’t replace actual flesh and blood. This is where the “serendipitous discovery” I mentioned earlier comes in. For both the marketer and the data scientist, creativity and the ability to actually know people—to be able to work with them on their level—can lead to discoveries a computer could miss.
If the work of data scientists goes towards replacing them with machines, we’ll see a very technologically advanced form of irony. Until that happens, we’ll have to be content with watching machines and humans deal in this most intangible resource of the modern age: big data.
