What the Future has in Store for Targeted Ads
As recent as a few decades ago, a lot of the tangible data we possess now was very abstract and almost completely impalpable. The idea of having hard data on the social network of 1/7th of the world’s population was unfathomable. Yet we’ve done it. The idea of GPS to solve our age old problem of how to get from A to B was thought to be impossible. Yet we’ve done it. Technology always exceeds our expectations, and information we’ve always thought was too large or too time-consuming to attain has become attainable.
Now, in modern day society, there are a countless number of networks we have attained or will attain in the future sometime. All these networks essentially involve the same components – us. From online social networks, to cash flow networks, to disease flow networks, telephone networks to even something as detailed as networks of diffusion of pharmaceuticals in doctor networks. Because all these networks have the same building blocks – people, we should be able to combine them into one gigantic network of information (Let’s be oblivious to privacy restrictions for a while). What might be the use of something on this scale, you might ask? The bigger the network, the more magic we can do with it.
Let’s take a look at some of the potential benefits of our gigantic networks –
1. Health –
Jennifer comes home after a long day of work. It’s late July in 2009, a few months before the flu pandemic is about to hit America. The network has, behind the scenes, noticed that a few people in North America, above average, have been purchasing flu medicine. One of them happens to be Sarah, Jessica’s friend from work. As Jessica walks over to her refrigerator, now equipped with a screen built just for ads, she sees that it recommends she get a flu shot because an epidemic was sensed. She trusts it, and gets a flu shot. Her neighbor, Melissa thus gets information about Jessica’s purchase and gets one herself, doing her part to prevent the spread of the flu through the social network.
As stated in the Friendship Paradox [3], your friends, on average are more likely to have more friends than you do. As a direct consequence, if you can trace phenomenon amongst your friends, you essentially centralize your position in the social network, and get access to information quickly. You therefore move yourself over to left of the Sigmoid Adoption Curve [4]. When it comes to detection of outbreaks of contagious diseases, this essentially means the ability to stop an epidemic.
The Blue Line shows the cumulative signature S-shaped or Sigmoid curve where the x axis denotes time and y axis denotes the the total number of people who have “adopted” the technology.
The yellow curve shows the density, or the number of people adopting it at that given time.
2. Targeting Ads focused on the Customer, not the Merchant
Brad hasn’t purchased detergent in a few months now and is running low. Being a busy banker in New York, he has better things to do than keep track of his detergent levels. As he turns on his TV after a long day at the office, the information network tells him he hasn’t made a detergent purchase in longer than the average time he usually does. This is followed by a list of targeted ads for detergent brands. Brad decides to press a button and receives his detergent at his doorstep the next day.
Because the gigantic network of all information works on the basic of human necessity, it shows itself precisely when the user would need it most. The advantage? Click through rates will sky rocket. The average individual has never clicked on a Google Ad, yet they seem to making more than $30 billion in revenue. When targeted with such precision, it helps both the user and the seller.
This illustration shows the click through rates of Google, which plummet quickly after the top few spots. Ad targeting of the future would have the same exponential plummet, but the click through rates in general would be much higher as the information backing the customer’s need is much greater.
3. Speeding up information flow
Jeremy has recently purchased a smartphone. He is incredibly happy with his purchase and the network senses he has spent over 15 hours of time using his phone in the week after his purchase. The next week, it sees that Jeremy went to buy a cover for his smartphone. Jeremy’s friend Jonathan isn’t very tech-savvy, but it seems that 3 or 4 of his distant friends, including Jeremy have really enjoyed using their new smartphone. An ad pops up, and Jonathan decides to purchase it.
Again, the establishment of a giant network means that you get information before you otherwise usually might, and from sources you might not otherwise interact with. This is essentially an act of steepening and heightening the Sigmoid Adoption Curve [4]. Essentially it means more profits, faster.
A graph demonstrating the steepening of the sigmoid adoption curve.
In conclusion, there are a lot of things which flow through networks that we are unaware of – smoking, obesity, and even emotion [5]. Lifestyles flow through network. All sorts of information as we know it flows through networks. What I’m trying to say is that the more networks we combine into one, more information flows through it, and faster. The ability to control this gigantic network would make possible things we could never do before. We could allow the flow of information we wanted to, and suppress the flow of information we didn’t through targeted advertising. The advertising of healthier food alternatives to obese people and calcium to osteoporosis patients, for example, would give us the power to better the world in our own way. We could predict and see trends, and stop the bad ones and supplement the beneficial ones. The two way benefit for the companies who advertise as well as the users the advertise to, and the increase in the efficiency of the advertisements are key to the future of targeted ads.
The clustering of overweight men and women in this graph demonstrates how even obesity, something one would normally not associate as contagious, flows through the social network.
This graph shows how even movie taste can flow through networks. Again the clustering of people who like the same movies signifies that it flows through the social network.
Sources:
- Ted Talk – Nicholas Christakis – How Social Networks Predict Epidemics
- Google Flu Trends in the US
- The Friendship Paradox
- Sigmoid Curve
- Ted Talk – Nicholas Christakis – The Hidden Influence of Social Networks