Online companies like Google and Facebook, among many others, make their money by tracking their users and selling this information for marketing purposes. How much is this data worth, per person, per year?
Because it’s a commodity that can be sold multiple times, and from which new data can be derived and sold, and because it can lead to sales and revenue that might not have otherwise existed, and because it doesn’t necessarily age out all that quickly (making it cumulatively valuable)… billions. Unmeasurable billions. More by the time you consider even the first few derivative layers of value and revenue created by its existence and use.
Anyone who (still) thinks this is irrelevant to them and their lives and economic well-being is invited to read the brand new The Aisles Have Eyes by Joseph Turow, one of the first books to put it all together in the sense that some of us have been discussing it here. There are more in press.
Anecdotally, a friend who works as a logistics manager for a large freight company said he’d give his right arm plus several other body parts for the sort of data that was routinely hoovered up from mobile phones and those sorts of sources.
He said if aggregated in the right way it would save the company millions per annum. His specific example was that if it could give you info to fine tune the location of a freight depot better by a few miles, that alone would save some vast amount of unwanted truck movements per annum.
Your job (maybe even whether you keep it). Your insurance rates. Whether you can buy a house our not. Your credit rating.
It’s about to get more valuable. And more voluminous. The House just voted to remove most privacy rules for ISP customers, and if it passes the Senate, Trump is guaranteed to sign it.
The important thing to understand is that “per person, per year” per the OP the value of the information is maybe a dollar. It is close to worthless … unless you can aggregate it into the largest possible bundles.
That’s why Google and Facebook and very few others have enormous valuations, even when they don’t make returns that are large by old-line corporate standards.
Your personal information is much more valuable to someone who steals your credit card number and puts a couple of thousand on it before you have a chance to cancel it than it is to any company on earth. But you can’t steal a billion peoples’ credit cards and use them.
And there is a flip side. If someone steals and uses your credit card you have a pure loss. If someone aggregates your information, they may make money on it. But you are almost guaranteed enormous value, in convenience, ease of use, product information, personal interactions, and all the other ways a wired society functions. People give up their information easily not merely because they are blinkered and stupid but because they recognize that they gain tremendously in virtually every virtual transaction.
Should consumers be educated about this process? By all means. Should the process stop? Over my dead body. The net benefits are too huge to lose.
I’ve been an advertiser on Google for many years and I seriously doubt that assumption.
Most people have no clue that Google’s core business is that they’re an advertising company that makes money by selling advertising space. They run a very sophisticated billboard company. Very sophisticated.
And it is fair to say that much of their success is founded on data and the clever analysis of that data. Google collects more data than you can possibly imagine and they use it to improve their business, not others.
Google makes some of the data available to webmasters via “Google Webmaster Tools” and “Google Analytics” and they make some of it available to advertisers via “Google Adsense” and “Google Adwords” and the “Adwords Keyword Tool” but I seriously doubt they’re selling off tracking data like you’re thinking.
It doesn’t make sense that they would. Why sell that valuable data to the Ornery Bob Advertising Agency and help me book advertisers, or sell it to General Motors and help them decide how to better run their Bing advertising campaign?
You’ll have to show me where I can buy this Google tracking data before I’ll believe it.
In the*** third quarter*** of 2015, Google’s advertising business generated $16.78 billion in revenue but the “per person” part of that is very difficult to know because not only are they tracking every visitor to Google (40K searches per second) they’re tracking every visitor to every web page connected to Google Analytics and I suspect most web pages are. Add to that the fact that most phone apps are calling web pages that aren’t public but are still often connected to Google Analytics and it gets unimaginably complicated real fast.
In other words, it ain’t that simple and only Google knows.
Excellent post!
I understand that if I go to Amazon with an idle curiosity about cat flaps, I will quickly see cat flap and associated merchandising on my FB page. I block ads on most sites and really don’t worry about it. Yes, I will quite probably get a few cat related offers by email, but they go straight to junk along with the fake lottery wins, offers of dubious drugs and nice people wanting to pay me to transfer their millions from some African country; I cull that several times a week.
That said, how does it impact me negatively?
Just saw this today:
Full article here
It already passed the Senate.
First, google up “reductio ad absurdum.” 
This is the level of dismissal this topic got a few years ago, when it was still semi-secret or at least not widely known how much consumer activity was tracked and monitored. It wasn’t then and isn’t now about simple matching your purchases to ads or product stocking; the other version of this dismissal is “Oh goody, it means my supermarket will always have my brand of peanut butter.” (Wrong, if your brand doesn’t fit the revenue value point.)
Put as briefly as possible, massive consumer tracking and data analysis means the whole economic net gets closed a lot more tightly around you. You don’t get more choices or your preferred choices; you get the choices discovered to bring maximum revenue from the target population. You can’t buy Grandma Jane’s peanut butter any more, because it generated 3% less profit per shelf foot than the median brand… but they know they won’t lose your business because you’ll just buy Jif instead (+6% in revenue generation). Multiply that across the spectrum for insurance, food, consumer products, cars, employment offerings and so forth - all optimized not to give you real choice, but to limit the populations choices to the smallest subset that generates the most profit.
And as much as that sounds like good ol’ capitalism and free market at work, it’s really something new and insidious in a number of ways. It means you will get what they think you will buy… and nothing else. More to the point, it means you will get what they pretty much know they can *make *you buy… and nothing else.
Put another way, it means that the consumer good, insurance, and even employment providers can essentially read your mind and thus win every hand of life poker - you will “win” fewer and fewer times in choice and benefit, getting only the smallest return (with the largest imagined set of “choice” and “benefit”) they can get you to accept. Think about it: an iPhone in nine colors is still just one product. Five variations of Jif is still just one product. Five companies offering car insurance plans with different names is just one choice. Ten companies hiring to identical slots is just one choice. And that’s the goal and purpose of massive data mining and big data analysis: to maximize revenue from the smallest possible set of consumer choices.
Choices in life alternatives, not just cat food.
That number seems totally out of line to me. A check of the AT&T 2016 Annual Report shows that their line item for consumer wireless service is $32 billion. This new data, calculated from an increase in cost, is worth two to three times their entire actual revenue? That can’t be true.
It’s simply a mistake to use consumer pricing levels to assume profits from another source. There is not a one-to-one matchup between them.
I’ve no doubt Congress gave the big companies a gift, but it’s not $30/month/household. Even $30/year/household would be a lot.
Also, it just cancelled protections that were supposed to go in place later this year. So nothing has really changed.
People keep posting while I’m posting. I can’t catch up. That’s the future that worries me.
I’m going to argue again that the offsetting benefits leave me with a net positive. The cereal aisle has grown since I was a kid, true, but it has hit a peak. (There are not two cereal aisles. In fact, in the largest supermarkets I know, it is just one side of one aisle in the front of the store.) In that aisle are far more choices that far better satisfy a far larger percentage of households than were happy when you could only get one variety of Cheerios instead of six or eight. That aisle also carries muesli and other variations on cereal that never used to be there. I see new variations all the time as well. If they sell well they stay permanently, a much more efficient process than that used to be. In another aisle there are other cereals, such as gluten-free ones. Gluten-freeness may be far overblown but it serves a very real segment of the population. I have parallel experience. I covered the rise of awareness of lactose intolerance and the number of specialty products that were lactose-free for three decades. That’s absolutely real. But no matter how hard companies pushed to move people onto lactose-free variants, only the core needy responded. The moderately needy didn’t perceive a net plus from doing so.
This is the counter-argument to the data-selling panic. The companies can lead consumers, they can tempt consumers, they can dangle myriads of products in front of consumers, but they have to offer not merely perceived value but net positive perceived value to the changes to stick. Nothing in data mining has changed this basic fact.
The idiot tech guru who ran JC Penney into the ground did so in the course of trying to turn a brick-and-mortar chain into a nationwide datamining operation, because the data from existing and new customers was deemed more valuable in the long run than selling a few smocks and sneakers.
That he destroyed a fine old chain with a solid niche is incidental. As a savvy friend pointed out, he failed at his intended purpose, too, despite being a tech ween and that everyone one else with a desktop computer and a copy of SPSS is finding ways to make money from consumer data processing. LoserLoser.
I think you’re making my point. A) He couldn’t make more money from data than by product sales. B) He couldn’t raise consumer spending because they perceived constant low pricing as a net loss from regular sales. That’s lose/lose but in exactly the way that fits what I’m saying.
To a certain point, and with a rather naive and trusting viewpoint, this is true. Certainly a fairly basic driver here is for companies to make the most they can from the least expenditures, and only rabid anti-capitalists (which I’m not) can object. There’s no reason a company should continue to produce a product that doesn’t pull its weight in the bottom line, directly or indirectly.
However, this is a sledgehammer tool that’s already being misused. Its scope goes far, far beyond determining what products show up on a grocery store shelf - that’s just the simplest and most at-hand example. When this tool starts being used to shape populations in ways other than purchase of consumer staples, it has the potential to generate control and influence on some much more important levels - including things like political choice.
And I completely disagree with the idea that consumers can’t be made to buy things or make choices; that sellers can only sort of pester us, not really influence us. The last century is absolutely filled with things that were hammered down the consuming public’s throat until they became “essentials.” Things that did not exist the week before and in the end serve no real useful purpose. The cycle of marketing products is built on creating niches and then filling them, and it’s been done uncounted thousands of times since WWII alone.
So when marketers (== consumer good multinationals, for the most part) start analyzing the population for what it can be made to buy, knowing with good precision what that thing is and how to pitch it so that the pitch bypasses consumer disinterest, lack of need and outright rejection filters… it’s not good for anyone but General Motors. Make no mistake, this is not just another iteration of “invent it and make them buy it.” It’s a new thing in this world, and it’s a terribly, terribly dangerous thing, as hints from the recent election tell.
Because the guy was a self-important fuckup.
His thousands of peers are succeeding just fine at both efforts, some more one way and some the other, and new entities join the game (mostly on the data side) daily.
Tell your friend to look into the AirSage data sets - this data exists and is available for purchase to a pretty high resolution. Sprint and Verizon aggregate the location data from their subscribers and make it available for purchase, and it includes things like travel patterns, dwell times, time series traffic in 15 minute increments, and more.