No free lunches
You’re being watched. You’re being tracked. You’re being profiled. You’re being measured. Your value is being weighed. Your every digital interaction leaves a data trail and these data trails are hoovered up by a dark web of surveillance companies. There’s a lot to hoover up. Much of Big Data is personal data.
Using sources from Every Day Big Data Statistics and Cisco, in 2016 Shaun Norris calculated that the average Internet user generates about half a gigabyte of data per day. That’s about 180 GB a year. According to Statista, “almost 4.48 billion people were active Internet users as of October 2019.” You’d need to plant almost 340 million trees to deal with the pollution caused by the personal data we generate on the Internet. It’s like when we’re online we’re belching CO2.
Frederike Kaltheuner, a tech policy fellow at Mozilla, has researched how much data was being collected on her. In just one week, a dark web (but totally legal) company called Quantcast “has amassed over 5,300 rows and more than 46 columns worth of data including URLs, time stamps, IP addresses, cookies IDs, browser information and much more.” Quantcast also had information on her gender, age, presence of children in her household (in number and ages), her education level, and gross yearly household income. All to target her with ads, though, as we’ll see later, this sort of targeting isn’t quite as effective as some dreamt it would be.
In the economy of free, the Earth pays. The economy of free is a hugely wasteful economy because it gives one million people a carrot in order to get one sucker to buy a juicer. The economy of free is a hugely wasteful economy because it is primarily an economy of wants rather than of needs. Why not download the app? It’s free. You may not use it more than once, but so what? It’s free.
I need a place to stay. I want to stay in a 5-star hotel at a 3-star price. The advertiser wants me to pay a 4-star price, or at least stay in their 5-star 3-star-price hotel. That requires a lot of targeting, a lot of persuasion, and all that requires a lot of energy. The energy to collect enough data on me to know that I am a good target. The energy to create highly emotional, visual advertising. The energy to track me and keep showing me this advertising wherever I go. That’s a lot of pollution.
The global online advertising industry is enormous. At over $300 billion in turnover, 2019 was the first year that more was spent advertising online than offline. ($300 billion, by the way, is the amount UN scientists estimate is needed to create sustainable solutions for climate change.) eMarketer estimates that online advertising will be worth more than $500 billion by 2023. Online advertising is horrendously wasteful, but that’s nothing new.
Offline advertising depends on “magic” and “creativity” and made-up numbers. Search-based advertising is much less wasteful than traditional advertising because the roles are essentially reversed. The searcher becomes the advertiser. When they search for “cheap flights Dublin,” they are advertising that they want to find a cheap flight to Dublin. It is then up to companies to respond to that ad with an offer of cheap flights to Dublin. That is a much more energy-efficient form of advertising because you’ve got a buyer with a demand that has been made clear (cheap flights Dublin) and a seller who can meet that demand.
As a result, it is estimated that for every 100 people that see a Google search ad, three will click on it. Three out of 100. That is regarded as a mind-bogglingly good metric by advertisers. Imagine if for every 100 people who walked into a restaurant only three stayed. Wouldn’t be very good, would it? But three out of 100 has made Google Google and Facebook Facebook. In the hugely wasteful world of advertising, three out of 100 is the most amazing of the most amazing success.
Now, let’s give it a little twist so as to bring traditional advertising into perspective. When was the last time you searched for “cheap flights new york” and ended up buying golf clubs instead, because you saw this amazing ad for golf clubs and it made you totally forget that what you really wanted to do was buy a cheap flight to New York? Doesn’t happen, does it? Happens in advertising magicland, at least that’s what the magical advertisers tell us.
When it comes to the more traditional visual banner ads, it takes 2,000 people to see such ads before one will click. This data is suspect, as so much data about the Web is, because it seems that 8% of people who do click account for 85% of clicks, comScore reported. Are these people “people” or are they bots masquerading as people? Or perhaps they are people but poor people working as click slaves, clicking and clicking on ads so as to defraud advertisers?
You see, just because people click on an ad doesn’t mean a lot. If you want to measure value, you must measure how many people actually bought the thing that was advertised in the ad. (And, as we’ll see later, you must know whether they bought it because they saw the ad, or whether they were going to buy it anyway whether they saw the ad or not.)
A 2015 study of 58 online ad campaigns, as reported by Science Direct, found that over 50% of them saw less than one purchase per million times the ad was seen. (This was considered an excellent result by ad executives.) Think about that for a moment. The ad was “seen” 999,999 times and no purchase occurred. All that pollution. Again, the economics are all wrong. The models we have today are designed to consume the Earth’s resources for short-term gain. The Earth pays for online advertising. Our children and grandchildren will pay for the wanton waste that is online advertising. If it’s free, the Earth’s paying.
According to Google, the maximum file size for a banner ad is 150 KB. Let’s take 100 KB as an average. If a typical banner needs to be seen one million times in order to achieve a purchase, then that’s 100,0000,000 KB, which is 100 GB. Transmitting one GB consumes about 0.015 kWh of electricity. For 100 GB that gives us 1.5 kWh which creates about 0.42 kg of CO2, according to RenSMART, for each banner ad over its lifetime.
Are these figures exactly correct? Maybe not. But what we do know with absolute certainty: It’s not zero. Every ad costs the planet. Every ad is based on a hugely wasteful model.
Online advertising costs you and me money as well, particularly if you are accessing the Internet with your smartphone using a data plan. For example, the cost of one GB of mobile broadband can be about $10 in the US and $7 in the UK. Prices vary a lot. The poorer you are, the more you tend to pay per GB because of the nature of the plans available. This is ironic because advertising is more and more targeted at the poor, since better-educated and better-off people have become more immune to ads. As economics professor Scott Galloway puts it, “Advertising is a tax the poor and technologically illiterate pay.”
It’s estimated that US citizens are exposed to thousands of ads every day, and about 55 of these are online banner ads. Thus, every day it costs a typical smartphone user in the US about six cents to download banner ads, or about $1.90 a month. It could in fact be much higher. In 2015, The New York Times, in a study of the mobile homepages of the top 50 news websites, found that more than half of all data transfers came from ads. The study estimated that for someone using a typical smartphone data plan, the ads alone on visiting Boston.com every day would cost them $9.50 a month.
What we’ve measured above is the cost of the file size of the ad. There are many more costs, not least the costs associated with designing and approving the ads. Apps with ads have been found to use an average of 16% more battery power, and sometimes up to 33% more, while consuming up to 50% of a phone’s processing power, because of all the tracking software that hangs off them. According to a 2018 study by Solarwinds, the average load time for the top 50 websites was 9.46?seconds with ads, and 2.69?seconds without. So, a website with ads took more than three times longer to download than one without. The same study found 298 pieces of tracking and personal data collection software on the top 50 websites. 75% of these trackers were ad-related.
In summary, online advertising increases energy consumption because:
- Much more data needs to be downloaded both for the ad itself and the huge amount of tracking software that comes with it.
- Much more processing needs to occur.
- The ads are constantly communicating across the network. Even when you’re not viewing the page, the ads are constantly sending bursts of energy-sapping data back to their motherlode in the data center.
Online advertising is full of fraud and malpractice. Data has shown that if you are Ford, for example, then buying ads when people search for “Ford” is not that effective because you’re going to reach those people anyway. However, if you are Ford and you buy ads to appear when someone searches for “Toyota” then that can be quite effective.
“Don’t be evil” has been the Google tagline, mission statement, guiding philosophy from practically day one of its existence. In 2018, it quietly dropped the “don’t” from the tagline. Google was entering a new phase of sucking up personal data and manipulating search results and spewing out giganormous profits.
Do you remember back in those innocent, hippy, don’t-be-evil days, when Google placed the ads in the right-hand column so as to clearly differentiate them from the organic search results? Ah, those naïve do-gooder, don’t-be-evil hippies that they were. We didn’t deserve them. Thing was, very few people looked in the right-hand column and even fewer people clicked. So, poor Google was only making billions from ads instead of the gazillions they dreamt of.
They quietly moved the ads to the top of the central column. And so that everyone would know that these were ads, they placed a teeny-weensy sign saying “Ad.” Such ethical dudes and dudesses. People are so stupid, though. Studies indicate that between 50% and 60% of people don’t actually know that the ads are ads. Google must be so upset about that.
That was just the beginning of nefarious activities for the rebranded “Be Evil” empire. Not satisfied with greying the line between ads and organic search results, Google decided to get into the highly lucrative Mafia Don business of shaking down companies for protection money. “When Google puts 4 paid ads ahead of the first organic result for your own brand name, you’re forced to pay up if you want to be found,” Jason Fried, founder of Basecamp states. “It’s a shakedown. It’s ransom.”
Like all good Mafiosi, Google targets vulnerable industry after vulnerable industry, looking for their protection money. Take travel. “TripAdvisor has one of the best link profiles of any commercially oriented website outside of perhaps Amazon.com,” search optimization expert Aaron Wall writes. “But ranking #1 doesn’t count for much if that #1 ranking is below the fold. Or, even worse, if Google literally hides the organic search results.” The result? Google grows its ad revenues 20% a year in a global economy growing at under 4%, and the stocks of Expedia and TripAdvisor fall off cliffs. That’s advertising for you.
It’s hard to deal with fraud when your industry is founded on it. Advertising so often depends on getting celebrities or “influencers” to peddle their wares. If brands can’t afford celebrities, then at least they’ll get beautiful happy people with amazing teeth to fake it.
Online, ad fraud has been taken to a new level. Why pay a celebrity to pretend they use your product when you can just steal their image. You can have them say exactly what you want for nothing. You can get them to claim that Product X has magical properties that will transform the life of all who buy it. How different is that from traditional advertising? Celebrities make wild claims all the time about products they probably don’t even use. When it comes to advertising, it’s hard to identify what is fraud and what is fake.
Julien has a popular app that depends on ads for revenue. “Julien sells a banner ad, which appears in the app and is visible to his users,” Craig Silverman writes for BuzzFeed. “Then, hidden from view behind that banner, fraudsters conceal autoplaying video ads that no human being actually sees, but which register as having been served and viewed. In this scenario, Julien gets paid for the small banner ad in his app that users see, but the fraudsters earn many times that amount by stuffing far more lucrative video ads behind the banner.” All those fraudulent video ads running in the background are pumping out loads of pollution.
It’s estimated that less than 60% of web traffic is human, with a large proportion of the rest being ad fraud bots clicking, clicking, clicking on ads. At one stage there were so many bots “watching” stuff on YouTube that management feared that their algorithms would start seeing the bots as normal and the humans as abnormal. Ad fraud costs billions every year but nobody really has a clue exactly how much, because when it comes to digital metrics, you’re entering a world of fakery, manipulation and magical thinking.
Facebook took down 2.2 billion fake accounts between January and March 2019, CNN reported. That’s an awful lot of fake accounts. Amazingly, only in 2019 did Facebook begin a proper auditing process for its ads. In 2016, it was estimated that 85% of “people” “watching” Facebook ads were “watching” with the sound off. Big fanbase for silent movies, I suppose. What was the solution? Captions for videos! In 2019, Facebook settled a legal complaint that stated that the average viewership metrics for video ads were inflated by between 150% and 900%.
Facebook, the company that pursues growth and profit over everything, has revolutionized ad fraud with its wonderful tools that can help criminals target the needy, weak and gullible. As one fraudster put it: “They go out and find the morons for me.” Advertising has a history of targeting the gullible, the weak, the poor, the uneducated, the needy. Online, it’s no different.
You might imagine that behind the scenes there is a big gang of evil geniuses figuring out the next way to trick and manipulate customers. You’d be wrong. Many organizations may be swimming in a sea of customer data but most of that sea is as polluted as our normal seas. “The majority of advertising companies feed their complex algorithms silos full of data even though the practice never delivers the desired result,” Jesse Frederik and Maurits Martijn wrote for The Correspondent.
Back around 2000, when a senior advertising executive was shown how Google worked, he wasn’t impressed. “You’re fucking with the magic” was his reply. In a different period, a manager from a brand was supposed to have said, “50% of my advertising works. I don’t know which 50%.” Google was promising to identify which 50%, and that was deeply worrying to traditional advertising executives, as transparency was the last thing they wanted in the advertising industry.
Advertisers pride themselves on being creative—as in “creative accounting.” They make shit up. Invent stuff to make their clients feel happy. They tell fabulous, magical lies to audiences that want to believe fabulous, magical lies.
For a while, Google did “fuck with the magic,” at least in relation to search-based advertising, which was much more transparent than traditional advertising had ever been. However, outside the realm of search-based advertising, online has turned out to be as big a world of magical thinking for advertising as the physical world was. For all the data, the metrics are still sprinkled with the magical dust of hype and buzzwords. In fact, if anything, data is overwhelming many organizations’ ability to be effective.
Although an incredible amount of personal data is collected, a great deal of it is never even used, because most organizations are not capable of managing data effectively. Instead, there is an obsession with big numbers and more volume. The volume metrics are often much more about how to make the higher-ups look good and feel good.
There is an obsession with meaningless concepts such as “engagement” and “interactivity,” “innovation” and creating things that go swish and swoosh, and making senior managers feel like they’re directors in a James Bond movie. Innovation is absolutely essential, whatever that means. When you boil it all down, so much of it is about the primal instinct to have more, to have a bigger whatever. It’s the Cult of Volume. Welcome to the Short-Term Growth Club.
Inventing high volume, meaningless metrics “makes everyone happy,” David Reiley, former head of Yahoo’s economics team explained to The Correspondent. “It will make the publisher happy. It will make the person who bought the media happy. It will make the boss of the person who bought the media happy. It will make the ad agency happy. Everybody can brag that they had a very successful campaign.”
Volume-based metrics have always been open to fraud and manipulation. The Irish police are responsible for saving lives on the roads by ensuring drunk drivers are found and prosecuted. That’s the desired outcome, but it was not the key metric for the police. The key volume metric that met targets and ensured promotions was how many drink-driving tests had been carried out. Oh, those big numbers, how management loves those big numbers. Of course, such input-focused, volume-based metrics are so much easier to manipulate. The Irish police kept inventing and inflating the huge and humongous number of tests they were carrying out, which showed to anyone that would listen what hard-working boys and girls they were.
Unfortunately, someone finally asked the very awkward question: “If you’re doing all these tests, how come you’ve so few convictions?” It’s much harder to manipulate data on the outcome (convictions for drunk driving). Focus on measuring the outcome if you want to establish true value.
Personalization was a wonder drug, an all-conquering magical tool that would personalize everything. Collect all this rich customer data and then micro-target the hell out of people. Sounded great in theory. However, to design and maintain such personalization systems takes a tremendous effort. I have seen personalization situations that required four to five times the investment in personnel than that required for running the same non-personalized website.
These people are not cheap. You need data analysts who can keep the data clean and accurate. You need designers, developers and content professionals continuously improving the environment, because personalization is notoriously difficult to optimize for.
Personalization is really, really hard work, fraught with privacy concerns, and at the end of the day often doesn’t deliver anywhere near the returns promised. Therefore, it was not surprising that in 2019 Gartner predicted that 80% of marketers would abandon their personalization efforts by 2025 “due to lack of ROI, the perils of customer data management or both.” The data, the promise of all that data, it wasn’t working the way the ads said it would. All that data collected for nothing. Waste. To be piled with all the other waste data in one more data dump in the Cloud.
What’s the answer? What’s the solution? Less data. Better quality. A much greater focus on customer outcomes, rather than organizational inputs. Much more analysis and thought.
Don’t feed the ads. Don’t click on the ad. Click on the natural search result instead. Because the more you click on the ads, the more they think you’re one of the gullible ones, and then they will relentlessly, relentlessly target you and sell you to others to relentlessly target.
Pay for stuff. Nothing is free. When it’s free, the Earth pays, you pay with bandwidth and your personal data is the product that’s sold to advertisers. Try it out for free, sure, but then have a serious, intense conversation with yourself. Is this app genuinely delivering value to my life? If yes, pay for it. You’re helping yourself, the environment and the app creator. You’re helping create a value-based economy rather than an ad-based economy.
If you make an app, sure, offer a free version, but if you can’t get people to pay, then question whether you are making something of genuine value, or are you one more waste creator, living off the advertising, living off the planet?
Just Do It. How about: Don’t Do It. Wait. Don’t download, don’t click, don’t buy. Do stuff with what you already have. Wear stuff out, use your stuff. Find ways to get what you need done with what you already have. Be inventive. Be creative. Be an innovator in your own life using the materials you already have. Weigh up everything you do. Is it worth it?
Don’t feed the ads. Don’t click on the ad. Find the organic search result and click on it. Organic is better.
Find out what you truly value. Pay for it.
World Wide Waste
- Free, weekly email
- Published since 1996
- Read some examples
- Subscribe with your email below
- How much digital data does an average digital user generate per day and per year? Shaun Norris, Quora, 2016
- RenSMART carbon emissions calculator
- Website Carbon Calculator
- How A Massive Facebook Scam Siphoned Millions Of Dollars From Unsuspecting Boomers, Craig Silverman, BuzzFeed, 2019
- Facebook removed 2.2 billion fake accounts in three months, Kaya Yurieff, CNN, 2019
- How Facebook Helps Shady Advertisers Pollute the Internet, Zeke Faux, Bloomberg, 2018
- Facebook Settles Class Action Claiming Company Inflated Video Viewership Metrics, Eriq Gardner, Hollywood Reporter, 2019
- These U.N. Climate Scientists Think They Can Halt Global Warming for $300 Billion. Here’s How, Adam Majendie, Pratik Parija, TIME, 2019
- How Much of the Internet Is Fake? Turns Out, a Lot of It, Actually, Max Read, New York Magazine, 2018
- The Effect of Ad Blockers on the Energy Consumption of Mobile Web Browsing, Arthur Visser, University of Twente, 2016
- After GDPR, The New York Times cut off ad exchanges in Europe — and kept growing ad revenue, Jessica Davies, DIGIDAY, 2019
- The Cost of Mobile Ads on 50 News Websites, G. Aisch, W. Andrews, J.Keller, New York Times, 2015
- Pixalate Says 30% Of Programmatic Ad Impressions In Australia Are Invalid… So Is Online Advertising A Scam? Edward Pollitt, Bandt, 2019
- Ads in free mobile apps have hidden costs for both users and developers, Robert Perkins, University of Southern California, 2015
- ‘A daily, hourly fight’: Digital ad fraud is worse than ever, January 9, 2019, Jessica Davies, Digiday, 2019
- The new dot com bubble is here: it’s called online advertising, Jesse Frederik, MauritsMartijn, The Correspondent, 2019
- The good, the bad and the troubling: trust in advertising hits record low, Gideon Spanier, Campaign, 2019
- The Cost Of Mobile Internet Around The World, Niall McCarthy, Forbes, 2019
- The 45 Most Important Advertising Statistics of 2019 (banner ad exposure), Milja Milenkovic, Smallbizgenius, 2019
- Gartner Predicts 80% of Marketers Will Abandon Personalization Efforts by 2025, Gartner, 2019
- I asked an online tracking company for all of my data and here’s what I found, Frederike Kaltheuner, Privacy International, 2018
- Almost 60% of People Still Don’t Recognise Google Paid Ads When They See Them, Varn, 2018