A few months back, an advertising executive argued that more online privacy would kill free speech.
Richard Frankel, president of the advertising company Rocket Fuel, titled the post “How the Do Not Track Plan Will Ultimately Kill Free Speech.”
It’s understandable that advertisers are so resistant to the concept of Do Not Track – it could change the way they access data, which would force them to innovate their business models and practices.
Just because DNT could impact the advertising industry in the short-term, however, doesn’t warrant its intense broadcasting of fear, uncertainty and doubt (FUD), and at times, outright deception. Frankels’s piece was one in a string of ad exec attacks on privacy that claim that the Internet will be doomed if consumers get more of the privacy they want.
In the interest of transparency, let me explain my angle here. I’m an attorney, privacy advocate, and analyst at Abine, an online privacy company in Boston. We make simple tools that give people a choice over whether their personal info is collected, stored, and sold online.
Unlike the advertising companies, we don’t collect or sell any user data. We only get paid if users like our free products and choose to buy our premium ones. It’s an up-front relationship that’s clear to our customers.
And beyond my role at Abine, I care deeply about preserving the web as a place where we can say and explore interesting, sometimes controversial, things. As a fan of free speech and expression, I have a vested interest in keeping the Internet open and uncensored.
Make no mistake about it: we live in a state of surveillance. Hundreds of advertising and tracking companies follow everything we do online -- the articles we read, the videos we watch, the sites we always visit, the Facebook comments we make, and more.
They combine that online data with offline data like our voting record, employment history, and marriage licenses, and use it to build an extremely detailed profile. Companies like Facebook scan the contents of photos and private messages for Homeland Security “risk words” like “infection,” “body scanner,” or “hacker” and turn them over to law enforcement.
Even if you delete your embarrassing Facebook posts, companies like Social Intelligence sell the past 7 years of posts to hiring managers. The wireless companies you pay for mobile service turn over 1.3 million customer records to law enforcement each year, which include texts and your phone’s GPS location wherever it went.
The divide between public and private surveillance is virtually nonexistent, and advertising companies are part of this ecosystem.
When you’re constantly being watched, you necessarily lack privacy. And when you’re constantly being watched, you act differently. That seems like an obvious point to those of us who’ve belted out song lyrics alone in our cars but would be terrified to do the same on stage in front of thousands or who laugh at things with their friends that they’d never say in front of their bosses, but observation effects are also well-established in scientific research, law, and popular culture (Big Brother, anyone?).
Privacy scholar Neil M. Richards writes that “surveillance inclines us to the mainstream and the boring…when we are watched while engaging in intellectual activities, broadly defined—thinking, reading, web-surfing, or private communications—we are deterred from engaging in thoughts or deeds that others might find deviant.”
Deviance just means difference: deviation from the norm, creativity, standing out from the crowd. If you knew you were being watched and that your activities may resurface down the road in a job interview, on a date, in a newspaper, you’d be less likely to go that political rally, that Alcoholics Anonymous meeting, that gay bar, or that protest, all things that you have a right to do that are valued parts of a free, democratic society.
With privacy, you have control over who gets to see what you’re doing and where you are. You get to be one way around your boss, another around your kids, another around your best friends since elementary school. You get to pick that divide.
But when you’re watched all the time and you never know how that information will be used or where it’ll turn up, you censor yourself everywhere. This is called the Panopticon Effect. All this openness that social networks insist we want by default, the pervasive data collection that advertisers argue is good for us: they make anyone who’s paying attention censor themselves.
A key part of free speech is anonymous speech. Anonymous speech is a constitutionally protected First Amendment right, and as the Supreme Court stated in McIntyre v. Ohio Elections Commission, “an honorable tradition of advocacy and of dissent.”
Online advertising constantly fights against anonymity and pseudonymity, trying to learn everything about web users to unmask them and peel away layers of demographic info, interests, and behaviors. Many social networks like Facebook have real name policies, insisting that members use their full, legal names on their accounts or be banned.
Frankel says that “With higher ad revenues comes more rich content that spurs vibrant discussion.” I disagree. You know what spurs vibrant discussion? People, especially people behind pseudonyms. Ever been on a message board? A Reddit thread?
The most “vibrant” discussions happen when people feel sufficiently protected to be honest. Sure, some people hide behind pseudonyms to harass others, but the core of the First Amendment avoids censoring the positive, protected uses of speech just because certain bad actors may abuse it.
To say that advertising is the only thing driving creative content doesn’t give credit to humanity’s ingrained creativity. First, plenty of content providers get paid for their work directly, from recording artists to bestselling authors to journalists. 43% of Americans pay content creators—authors of magazine articles and books—for e-books, and there are 400,000 paid subscribers to the Wall Street Journal alone.
The majority of people blogging and posting on social media today create content out of a desire to express themselves that’s unrelated to money. Just talk to any one of the thousands of bloggers and podcasters, particularly the anonymous ones, who pay to host their own websites just to get their message out to the public.
Most of these people don’t blog or podcast for a living; they do it because they love it and because their message is important to them. There’s power—and satisfaction—in speaking one’s mind, and it exists independently from ad money.
Privacy isn’t a hindrance to free speech; it’s the driving force behind it. Privacy provides both the boundaries of and protection for the space in which we can be ourselves. Privacy nurtures self-expression, creativity, speaking your mind, associating with whomever you wish, and exploring your interests.
These are the First Amendment’s protections: freedom of speech, of association, and of assembly. They’re so important for self-actualization and self-determination that our founders immortalized them in the Bill of Rights. Privacy isn’t about having something to hide; it’s about having something to live for.
Why the ad industry is wrong about Do Not Track
A few months back, an advertising executive argued that more online privacy would kill free speech. Richard Frankel, president of the advertising company Rocket Fuel, titled the post “How the Do Not Track Plan Will Ultimately Kill Free Speech.”
It’s understandable that advertisers are so resistant to the concept of Do Not Track – it could change the way they access data, which would force them to innovate their business models and practices. Just because DNT could impact the advertising industry in the short-term, however, doesn’t warrant its intense broadcasting of fear, uncertainty and doubt (FUD), and at times, outright deception.
Frankels’s piece was one in an ongoing string of ad exec attacks on privacy that claim that the Internet will be doomed if consumers get more of the privacy they want.
In the interest of transparency, let me explain my angle here. I’m an attorney, privacy advocate, and analyst at Abine, an online privacy company in Boston. We make simple tools that give people a choice over whether their personal info is collected, stored, and sold online. Unlike the advertising companies, we don’t collect or sell any user data.
We only get paid if users like our free products and choose to buy our premium ones. It’s an up-front relationship that’s clear to our customers. And beyond my role at Abine, I care deeply about preserving the web as a place where we can say and explore interesting, sometimes controversial, things.
As a fan of free speech and expression, I have a vested interest in keeping the Internet open and uncensored.
Make no mistake about it: we live in a state of surveillance. Hundreds of advertising and tracking companies follow everything we do online -- the articles we read, the videos we watch, the sites we always visit, the Facebook comments we make, and more.
They combine that online data with offline data like our voting record, employment history, and marriage licenses, and use it to build an extremely detailed profile. The divide between public and private surveillance is virtually nonexistent, and advertising companies are part of this ecosystem.
In this post, I’ll address Frankel’s argument - and the other advertisers using similar arguments - which goes something like this: 1) behavioral (or tracking-based) advertising is how online advertisers make money; 2) without advertising, content providers like online newspapers won’t be able to survive; 3) with paid subscriptions; only rich people who can afford content will get it, creating class segmentation.
In brief, here’s why they’re wrong:
1) the majority of online advertising revenue comes from contextual ads, which don’t pose any personal privacy risks;
2), the online advertising industry’s move to real-time bidding platforms are actually hurting content providers far more than a shift from behavioral advertising to contextual advertising ever could; and
3), beyond this being a reductio ad absurdum argument (“everything will be locked behind a paywall if we stop tracking!”), this segmentation already exists through the filter bubble that the advertising industry and its tracking has caused.
Realistically, online privacy isn’t going to hurt anything but these advertisers’ antiquated business models, which will have to adapt to respect the privacy of their target audiences.
Let’s first examine the fallacy that online tracking doesn’t use personal information. Frankel argues that “Targeted, relevant ads don’t harm consumers—personal information is not necessary, or collected, to produce them.”
This statement is less than fully honest when we look at Rocket Fuel’s own website, which boasts that it can “zero in on” people based on “age, gender, profession, ethnicity, and relationship status,” among many other personal characteristics. They have more than 20,000 audience segments, including soccer moms and caregivers:
The company goes on to say that they can find “users engaged in highly targeted activities that define their interests and personalities” by going “beyond behavioral, contextual, or geo-targeting by combining intelligent demographic, lifestyle, purchase-intent, and social data with our own suite of targeting algorithms, blended analytics, and expert analysis.”
And they’re not just using your online activity here: “Online data has evolved from simply providing insights like browsing activity, shopping cart info, sign ups, etc. into an incredible selection of more than 20,000 unique audience segments. Advertisers can even leverage offline purchase and consumption data for their online campaigns.”
Obviously, significant amounts of personal information are being collected. And the traditional advertiser counter-argument that “it’s not personally identifiable; it’s only aggregate info” has been debunked time and time again by researchers like Stanford’s Arvind Narayanan and Jonathan Mayer (also see privacy professor Paul Ohm’s excellent summary of the failure of anonymization) and publications like the Wall Street Journal, which found in a December 2012 study that nearly 25% of the web’s 70 most popular sites shared personal data, like name and email address, with third-party companies.
Ad companies like Mixpanel come right out and say “Now you can tie any kind of data to your users to see exactly who they are and what they have done.” The myth that de-identified data is private is even weaker now with the rise of “data enhancement” that matches online info with offline data sets.
The only reason these ads can be personalized is because of the personal information, the data collection, that powers them. That’s the real harm here, and that’s always been the harm, despite advertisers trying to shift the focus to targeted ads. No one cares about targeted ads: at worst, they’re annoying or creepy. Let’s just drop it.
The thing people don’t like is having their personal info harvested, mined, sold, and used in ways they can’t even imagine: for determining their credit limits or creditworthiness, playing into whether they get a job, showing them different prices than other people see for the same online shopping items, or influencing their insurance rates.
Even Frankel seems to doubt himself: he says that “consumers have become more open to [ads],” and then contradicts himself 6 paragraphs later when he says “Everyone claims to hate online advertising.” His second statement is more accurate: an October 2012 UC Berkeley study found that most people--69%--never or hardly ever find ads useful, and 85% never or hardly ever click them; and a 2012 Pew phone survey of 2,253 participants found that 68% of them were “not okay with targeted advertising because [they] don’t like having [their] online behavior tracked and analyzed.”
Microsoft surveyed their users about privacy in 2012 and found that 83% of Americans, 84% of Germans, and 81% of UK residents think that “tracking of personal information is out of hand and consumers need easier ways to block it.” I won’t keep going; you get the picture. The advertising industry keeps preaching about how great data collection is for all of us. If the benefits are so clear, then why not let consumers choose to enable it? Overwhelmingly, the opposite is happening.
Let’s fact-check some of Frankel’s, and the advertising industry in general’s, more unsubstantiated statements:
Online ads have become “less invasive”
If by “invasive” he means “in your face” like pop-ups, then he’s right. But just because data collection is invisible doesn’t mean it’s somehow safer or less intrusive. Advertisers employ a massive array of secret tracking techniques, collecting far more personal information than ever before in history, and they haven’t had the best rap while doing it: remember KISSmetrics creating undeletable cookies?
Google circumventing privacy controls in Safari and getting a $22.5 million FTC fine? Ad network Chitika’s opt-out that only lasted 10 days in contrast to any reasonable consumer’s expectations? Facebook tracking logged-out users anywhere there’s a Like button on the web?
Online tracking is only getting worse: UC Berkeley’s Web Privacy Census, powered by Abine’s privacy software, found an alarming increase in tracking on the top US websites, showing that online tracking will double in 2.5 years if present trends continue.
“The alternative to an ad-supported Internet is a pay-for-play world supported by subscriptions or private ownership. Consumers may think they want an ad-free Internet, but do we really want to pay subscription fees to access all the sites we currently visit for free?”
Frankel wrongly equates “behavioral advertising” with “all advertising” and offers a false choice between an ad-supported Internet and an all-paid subscription Internet. He neglects to mention that the vast majority of advertising revenue comes from contextual ads, ads that relate to the content the viewer sees and not the personal characteristics of the viewer.
Contextual ads don’t present a privacy problem, and they make up the majority of online advertisements. The reality is that privacy controls will not have a negative impact on the economics of the Internet (at least not past this short-term period of transition), as I’ve argued elsewhere.
And it’s not as though advertisers are truthfully concerned about publishers’ bottom lines. Significant evidence suggests the opposite: although online advertising is getting cheaper for advertisers, it’s getting more costly for publishers.
In the past, publishers had the power when pricing ads. You wanted a full-page New York Times ad? There’s a set price for that. With the recent advent of real-time online bidding platforms, advertisers hurt content providers by competing more effectively and cutting down the content provider’s take and involvement.
And at bottom, there’s plenty of research showing that plenty of people will pay small price increases for more privacy. A few dollars is all that’s needed to make the difference, so we’re not talking about a prohibitive expense. Offer people a choice: those who don’t mind surrendering their privacy can keep the status quo, but people who want to avoid it can throw in a few bucks to do so.
The Do Not Track option could serve as the dividing line to signal a consumer’s choice about privacy. Of course, it would need to live up to consumer expectations and actually stop tracking on websites, which notable websites like Twitter have done.
Do Not Track would “reduc[e] the effectiveness of advertising”
To the contrary. Do Not Track has great potential to increase the effectiveness of advertising. Why? Because Do Not Track lets people tell advertisers not to use certain personalized ads that they find ineffective, usually because they find them creepy or annoying.
Do Not Track is a clear communication from web users to advertisers about which ads those users prefer; which ads work better for them. If ads annoy people and make them build negative associates with the things advertised, they’re not effective advertising.
Even with Do Not Track, advertisers still get to collect and sell user data for advertising: they just can’t show personalized online ads. According to the definition of Do Not Track offered by advertisers (which is at odds with what real people want), people who don’t want to be tracked will still see ads, but not personalized ones. Meanwhile, all other web users will keep seeing behaviorally targeted ads.
Across the board—regardless of whether a consumer clicked Do Not Track—advertisers will still collect all consumers’ personal data exactly as before. Again, their definition, not ours. This definition means that consumers get to voice which type of ads they prefer—personalized or not—and advertisers can still collect and sell everyone’s info. It’s a win-win for advertisers, far from the end of the Internet economy and a blow to advertising effectiveness.
“If independent publishers lose a large percentage of their ad revenues because Do Not Track or other initiatives restrict the free flow of information on the web, they’ll have less money to fund unbiased, journalistic content creation”
We already have a restricted flow of free information on the web, thanks to advertising. Ever heard of a filter bubble? It’s “a situation in which a website algorithm selectively guesses what information a user would like to see based on information about the user (such as location, past click behavior and search history) and, as a result, users become separated from information that disagrees with their viewpoints, effectively isolating them in their own cultural or ideological bubbles.”
In other words, the more that advertisers or websites (or both, in the case of sites like Google and Facebook) know about you, the more they envelop you with targeted content. You don’t see the same Google search results as everyone else, or the same Amazon home page. Even news sites are targeted to show you articles they think you’ll like. Personal data collection fuels filter bubbles.
Frankel says that greater privacy will “muffle the voices of many consumers who can’t or won’t pay to express their opinions.” You know what muffles voices? Surveillance, tracking, and the threat that what you say today will be used against you in the future.
That’s why anonymous speech is constitutionally protected and why many people choose to mask themselves when they make political comments. Online advertising works extremely hard to unmask those people, identify them, sell their data, and barrage them with ads. And that’s not what people want.
Please follow Advertising on Twitter and Facebook.