Many people realize that smartphones track their locations. But what if you actively turn off location services, haven’t used any apps, and haven’t even inserted a carrier SIM card?
Even if you take all of those precautions, phones running Android software gather data about your location and send it back to Google when they’re connected to the internet, a Quartz investigation has revealed.
Since the beginning of 2017, Android phones have been collecting the addresses of nearby cellular towers—even when location services are disabled—and sending that data back to Google. The result is that Google, the unit of Alphabet behind Android, has access to data about individuals’ locations and their movements that go far beyond a reasonable consumer expectation of privacy.
Quartz observed the data collection occur and contacted Google, which confirmed the practice.
The cell tower addresses have been included in information sent to the system Google uses to manage push notifications and messages on Android phones for the past 11 months, according to a Google spokesperson. They were never used or stored, the spokesperson said, and the company is now taking steps to end the practice after being contacted by Quartz. By the end of November, the company said, Android phones will no longer send cell-tower location data to Google, at least as part of this particular service, which consumers cannot disable.
“In January of this year, we began looking into using Cell ID codes as an additional signal to further improve the speed and performance of message delivery,” the Google spokesperson said in an email. “However, we never incorporated Cell ID into our network sync system, so that data was immediately discarded, and we updated it to no longer request Cell ID.”
A cell-tower location sent to Google from an Android device. (Obtained by Quartz)
It is not clear how cell-tower addresses, transmitted as a data string that identifies a specific cell tower, could have been used to improve message delivery. But the privacy implications of the covert location-sharing practice are plain. While information about a single cell tower can only offer an approximation of where a mobile device actually is, multiple towers can be used to triangulate its location to within about a quarter-mile radius, or to a more exact pinpoint in urban areas, where cell towers are closer together.
The practice is troubling for people who’d prefer they weren’t tracked, especially for those such as law-enforcement officials or victims of domestic abuse who turn off location services thinking they’re fully concealing their whereabouts. Although the data sent to Google is encrypted, it could potentially be sent to a third party if the phone had been compromised with spyware or other methods of hacking. Each phone has a unique ID number, with which the location data can be associated.
The revelation comes as Google and other internet companies are under fire from lawmakers and regulators, including for the extent to which they vacuum up data about users. Such personal data, ranging from users’ political views to their purchase histories to their locations, are foundational to the business successes of companies like Facebook and Alphabet, built on targeted advertising and personalization and together valued at over $1.2 trillion by investors.
Cell-tower locations collected and sent to Google from an Android phone with location services turned off and carried in Washington, DC. (Obtained by Quartz)
The location-sharing practice does not appear to be limited to any particular type of Android phone or tablet; Google was apparently collecting cell tower data from all modern Android devices before being contacted by Quartz. A source familiar with the matter said the cell tower addresses were being sent to Google after a change in early 2017 to the Firebase Cloud Messaging service, which is owned by Google and runs on Android phones by default.
Even devices that had been reset to factory default settings and apps, with location services disabled, were observed by Quartz sending nearby cell-tower addresses to Google. Devices with a cellular data or WiFi connection appear to send the data to Google each time they come within range of a new cell tower. When Android devices are connected to a WiFi network, they will send the tower addresses to Google even if they don’t have SIM cards installed.
“It has pretty concerning implications,” said Bill Budington, a software engineer who works for the Electronic Frontier Foundation, a nonprofit organization that advocates for digital privacy. “You can kind of envision any number of circumstances where that could be extremely sensitive information that puts a person at risk.”
The section of Google’s privacy policy that covers location sharing says the company will collect location information from devices that use its services, but does not indicate whether it will collect data from Android devices when location services are disabled:
When you use Google services, we may collect and process information about your actual location. We use various technologies to determine location, including IP address, GPS, and other sensors that may, for example, provide Google with information on nearby devices, Wi-Fi access points and cell towers.
According to the Google spokesperson, the company’s system that controls its push notifications and messages is “distinctly separate from Location Services, which provide a device’s location to apps.” Android devices never offered consumers a way to opt out of the collection of cell tower data.
“It is really a mystery as to why this is not optional,” said Matthew Hickey, a security expert and researcher at Hacker House, a security firm based in London. “It seems quite intrusive for Google to be collecting such information that is only relevant to carrier networks when there are no SIM card or enabled services.”
While Google says it doesn’t use the location data it collects using this service, its does allow advertisers to target consumers using location data, an approach that has obvious commercial value. The company can tell using precise location tracking, for example, whether an individual with an Android phone or running Google apps has set foot in a specific store, and use that to target the advertising a user subsequently sees.
After a slow start out of the gate, adoption of ads.txt is taking off with some help from Google.
Only 13 percent of the 10,000 most popular domains that sell digital ads adopted ads.txt in the 100 days following its release by the Interactive Advertising Bureau Tech Lab in May. But between mid-September and the end of October, that number jumped to 44 percent, according to data from Ad Ops Insider publisher Ben Kneen. The time frame of the uptick in adoption overlaps with Google’s recent announcements that several of its most popular ad products will begin filtering for ads.txt. It also underscores just how much sway Google holds over publishers.
The IAB Tech Lab launched ads.txt as a tool to help ad buyers avoid illegitimate sellers that arbitrage inventory and spoof domains. The way that ads.txt works is that publishers drop a text file on their web servers that lists all the companies authorized to sell their inventory, which allows buyers to check the validity of the inventory they purchase. If a publisher uses ads.txt, then anyone with an internet connection can verify the publisher’s authorized sellers.
Domain spoofing remains a serious threat to the growth of programmatic advertising, so aside from shady resellers, pretty much everyone who works with automated advertising has incentive to back ads.txt. But despite the fanfare that followed ads.txt’s announcement, publishers dragged their feet in adopting ads.txt early on. The text files take just a few hours to create, according to multiple publishing sources, but publishers were slow to adopt because they had finite tech resources overcommitted to other projects, a lack of understanding about how ads.txt could benefit them and a lack of clarity about the impact. (Some publishers also weren’t crazy about exposing all the ways buyers could get onto their site outside of direct sales.)
“Even one or two months ago, it was slow going,” said Dennis Buchheim, svp and gm of the IAB Tech Lab, noting that ads.txt has since become one of the Tech Lab’s fastest-growing initiatives. “People were saying good things, but not necessarily acting on it.”
Google isn’t alone in pushing for ads.txt, so it can be difficult to separate correlation from causation when trying to determine what prompted so many publishers to create ads.txt files recently. Some ad buyers have said they won’t buy from publishers without ads.txt by the end of year, ad industry lobbyists support ads.txt and ad tech firms like AppNexus and MediaMath have created tools around the initiative. A few publisher execs have also clamored for more adoption of ads.txt to protect media brands from fraudsters, and ads.txt has become a buzzword at industry events.
All of these factors likely helped goose adoption rates, but none of these other organizations have Google’s power. Before Google announced it would use ads.txt files to filter unauthorized inventory, just 20 percent of the publishers that work with ad tech firm Advelvet had ads.txt files, said Murat Deligoz, CEO of Advelvet, which helps publishers set price floors in Google’s ad exchange. Now, all the publishers that work directly with Deligoz’s company have ads.txt files.
“Google is definitely the force to get ads.txt up and running,” he said.
In the past six weeks, Google announced that its DoubleClick Ad Exchange, AdSense network and demand-side platform DoubleClick Bid Manager will begin using ads.txt filesto filter unauthorized inventory. For now, publishers without an ads.txt file can still sell their inventory through these channels. But if a publisher does have an ads.txt file, Google will block unauthorized vendors not listed in the file from selling the publisher’s inventory, said Pooja Kapoor, Google’s head of programmatic global strategy.
“I believe that Google should get most of the credit [for the increase in ads.txt adoption] because until the DoubleClick Bid Manager mandate went out a few weeks ago, many publishers, though not all, were just playing a waiting game,” said Michael Stoeckel, vp of global tech strategy at Prohaska Consulting.
Aside from the public announcements, Google has worked in the background to push ads.txt. In mid-October, Google sent emails to publishers using its DoubleClick exchange that said publishers should update their ads.txt files “in order to prevent impact to your earnings.” In the past month, Google’s ad server added an ads.txt management tab to its dashboard that shows publishers which sellers are listing their domains without authorization, according to a publishing source requesting anonymity.
Screenshot of the ads.txt tab in Google’s ad server
Last week, Google launched a tool that scans publishers’ ad calls and spits out ads.txt files for them based on who it thinks is authorized to sell their inventory, Kapoor said. Publishers are encouraged to fact-check and tweak these files. The idea isn’t to set a publisher’s ads.txt file in stone but instead to remove friction to help it start one.
Google’s recent moves regarding ads.txt can be seen as the first steps that ultimately culminate with Google requiring ads.txt files from publishers before they can sell inventory through Google platforms. Google has discussed requiring ads.txt files from publishers, but no timeline exists for making such a move since ad buyers and sellers are just beginning to account for ads.txt in their transactions, Kapoor said.
“There’s no reason why we as an industry should monetize counterfeit inventory,” Kapoor said.
Since Google owns the most popular ad exchange, DSP and ad server in the ad industry, it wants to prevent further erosion in the trust in programmatic. Google’s ad exchange is listed on nearly 14,000 ads.txt files, which is 50 percent more files than the second-most popular exchange AppNexus has, according to ad measurement company Pixalate. Propping up trust through initiatives like ads.txt is one way for the search giant to encourage ad buyers to keep sending dollars through programmatic platforms, which is an area Google dominates.
Although publishers and advertisers alike have wearied of the power that platforms like Google and Facebook possess, the industry is welcoming Google’s push for ads.txt adoption. Liane Nadeau, director of programmatic at ad agency DigitasLBi, called Google’s backing of ads.txt “a great step.”
Slate adopted ads.txt in September a few weeks before Google announced its DSP would start using ads.txt files to filter unauthorized sellers, said Dan Check, vice chairman of The Slate Group. While Google wasn’t the main catalyst to nudge Slate to use ads.txt, Check recognized and supported Google’s use of its influence to push for industrywide ads.txt adoption.
“It can be good to have a stick, waved judiciously, in cases like this,” he said.
Silicon Valley’s online tech giants are under pressure, and not just from Russia investigators. This time it is the people who pay the bills: the advertisers who are asking serious questions about whether their products were sold alongside covert Russian propaganda.
Just as airlines pull their ads during coverage of air crashes, advertisers have long had strict rules about the placement of their brands. But the evolution of the automated digital ad business, including that of Facebook, Twitter and Google, has led to some distasteful situations for advertisers, including the placement of hundreds of their banner ads — including those for politicians — atop jihadi videos on YouTube.
That debacle has led advertisers to be especially sensitive to what they call the issue of “brand safety,” the effort to make sure their commercials appear next to quality content.
The sun rises behind the sign at the entrance to Facebook headquarters in Menlo Park before the company’s IPO launch in 2012. Beck Diefenbach / Reuters file
“The controls are not as strong as one would like,” said Bill Koenigsberg, chief executive of Horizon Media, an ad agency that spends around $8 billion a year for clients from Geico to Burger King. “Facebook and Google have hired people to be brand-safety watchdogs. Marketers are asking a lot more questions, the agencies that represent them are trying to hold Facebook and Google a lot more accountable, and if they don’t clean up the garden you will start to see it affect their pocketbooks.”
He said it isn’t just the Russia investigations that are causing concern but the parade of bad news surrounding the online advertising business. “Every day there is another snippet that comes out, and you piece them together and you’re starting to draw conclusions that the water isn’t as clear as we’d hoped.”
Facebook said it will place 1,000 staff members on a team to review ads and soup up its machine-learning abilities to address the problem. But the revelations are dragging the company and other online firms into a huge and potentially costly embarrassment.
No companies have pulled their ads from Facebook yet, at least not publicly, but executives say the sentiment is turning negative. That could have substantial implications for the company’s bottom line, which is almost entirely dependent on ad revenue.
One high-level business strategist, Michael Kassan, chief executive of Medialink, which advises tech companies and large media conglomerates, said that chief marketing officers at big companies fear that their bosses will ask them what is adjacent to their ads.
“People are on much higher alert,” Kassan said. “I have never seen it so pronounced. It’s every one of them.”
Kassan thinks the Russia investigation is still too new for marketers to figure out how it affects them.
“But you can’t argue they’re not involved,” he said. “If your brand is supporting the inclusion of stuff you don’t think is appropriate and you have proximity to it? One is known by the company one keeps.”
Nonetheless, there is still a clear reluctance in the advertising world to publicly criticize Facebook and Google.
“The entire advertising world is very anxious,” said Mike Paul, an independent expert in crisis public relations, who has worked at big ad agencies, “but few will admit publicly that the negative news is affecting Facebook because it is the 800-pound gorilla globally for ad and media buyers.”
Facebook’s Credibility Problem
The nervousness is not just because of the Russia inquiry. Brian Wieser, an analyst with Pivotal Research, recently found that the company claimed to reach an audience of 41 million people in the U.S. ages 18 to 24, though only 31 million of them exist, according to the Census Bureau.
The Video Advertising Bureau, an organization of broadcasters and cable companies, issued a study on Oct. 2, asserting that there are fewer people ages 18-34 living in every state than the number of people in that age range that Facebook claims it can reach in those states. (NBCUniversal, the parent company of NBC News, is a member of the bureau, as are the other major networks.)
Facebook said the problem may have been caused by kids pretending to be older, and said it does not bill advertisers on the basis of those estimates.
Facebook bought a series of newspaper ads, which ran on Wednesday, to explain its commitment to transparency. “We take the trust of the Facebook community seriously,” the ads said. “We will fight any attempt to interfere with elections or civic engagement on Facebook.”
Last month, ProPublica, the investigative news site, revealed that Facebook advertisers were able to use self-serve tools to target “Jew haters.” The company’s chief operating officer, Sheryl Sandberg, acknowledged the failure.
“We never intended or anticipated this functionality being used this way – and that is on us,” she said in a company statement.
A few days later, Mark Zuckerberg, Facebook’s chief executive, admitted that he was wrong to dismiss fake news on the social network as having had any influence on the election. “Calling that crazy was dismissive and I regret it,” he said in a statement.
Facebook and Google: The “Digital Duopoly”
Facebook is the second-biggest player in the digital ad market after Google. Both companies are set to grow their share of the ad pie. Together they are often known as the “digital duopoly,” with the kind of global scale other media companies can only dream of. Facebook has 2 billion monthly users. YouTube, owned by Google, has 1.5 billion.
And the financials are staggering by any measure.
This year, Google (including YouTube) will garner $35 billion in total digital ad dollars in the U.S., up 18.9 percent from last year, according to eMarketer, a measurement firm. That expansion will push Google’s share of the U.S. digital ad market to 42.2 percent.
The firm forecasts Facebook’s total digital revenues in the U.S. will grow 40.4 percent to $17.37 billion, pushing its share of all U.S. digital ad business to 20.9 percent.
Mark Zuckerberg did a Facebook live to discuss Russian election interference and next steps to protect the integrity of the democratic process. Facebook
Advertisers have to take Facebook and Google’s word that they get what they pay for. “We do not have visibility into whether we got what we bought,” said a senior marketing executive with a Fortune 500 company, who added: “I’m really nervous. We’re in business with Facebook and Google, they are the gateway to the audience, but the relationships are strained.”
Meanwhile, Facebook is on track to have yet another killer quarter. Last week a Deutsche Bank analyst, Lloyd Walmsley, told investors: “Facebook is the new IBM (in a good way). …We think Facebook is growing into a similar position in advertising, with best-in-class ad systems, a large growing audience across numerous products and a well-oiled sale machine.”
YouTube’s Credibility Problem
Still, Alphabet’s Google — is facing its own YouTube drama with advertisers around the globe.
One ad executive, who did not want to be identified because of business relationships with the online companies, said that a recent test run of ads on YouTube found they were placed adjacent to content that didn’t meet the firm’s requirements 30 percent of the time. After the Las Vegas shooting last week, for example, false conspiracy theories wound up high on YouTube’s search results, as reported in the Wall Street Journal. YouTube said it would tweak its search results to show more reliable sources of news.
Some of YouTube’s largest advertisers in 2016 dropped their spending by 95 percent or more, according to Pathmatics, a company that tracks online advertising, and measured desktop ad spending.
The company says AT&T notably reduced spending on YouTube by 76 percent over the year before (January through August). Disney spent less than $200,000 from April to August after spending more than $1 million in January, the data showed. Overall, however, Pathmatics said that spending on desktop YouTube was up 31 percent for the period January through August versus the same period last year.
In August, Google said it would refund advertisers for ads placed on dodgy websites with fraudulent traffic counts, according to a Wall Street Journal report.
Proctor & Gamble, the packaged-goods giant, cut more than $100 million in digital spending beginning in March, the company’s chief brand officer, Marc Pritchard, said in a speech in Orlando, Florida, on Thursday.
“There is no question ads should ever be on an ISIS video,” Pritchard said.
A Wake-Up Call For the Industry
Martin Sorrell, chief executive of the WPP Group, an advertising holding company whose agencies spend billions of ad dollars around the globe, said the pressure on Facebook is going to intensify. He said he has urged Facebook for some time to acknowledge that it is not just a platform.
“They are a media company,” he said. “They seem to have acknowledged the need for human review and that you can’t just rule by algorithm alone.”
Randall Rothenberg, president and chief executive of the Interactive Advertising Bureau, which represents Google, Facebook and other big content companies and advertising firms, said there’s an industry-wide effort to fix a host of issues, from fraud to cybersecurity to fake news.
“What has changed over the past three to four months, thanks in no small part to the Mueller investigation and the dribbling of information from various parties, is a broader understanding from multiple constituents that these things are all connected,” he said, referring to Robert Mueller, the special counsel investigating Russian meddling in last year’s election.
“The common understanding is out there,” Rothenberg added, “and now there is a greater will among more parties to come to the table to solve it.”
People are freaked out by ads that follow them around after a google of the product.
By Mediastreet Staff Writers
Personalised ads now follow us around the web, their content drawn from tracking our online activity. We in the ad industry have suggested that people are okay with it – that people see benefits roughly equal to perceived risks.
A study by University of Illinois advertising professor Chang-Dae Ham says otherwise, suggesting the ad industry may want to reconsider its approach.
“The perception of risk is much stronger than the perception of benefit,” Ham found in surveying 442 college students on how they coped with what is known as online behavioural advertising. “That drives them to perceive more privacy concern, and finally to avoid the advertising,” he said.
Previous studies have looked at various aspects of online behavioural advertising (OBA), but Ham said his is the first to investigate the interaction of various psychological factors – or mediating variables – behind how people respond to it and why they might avoid ads.
“The response to OBA is very complicated,” he said. “The ad avoidance is not explained just by one or two factors; I’m arguing here that five or six factors are influencing together.”
Ham examined not only interactions related to risk, benefit and privacy, but also self-efficacy (sense of control); reactance (reaction against perceived restrictions on freedom); and the perceived personalization of the ads.
He also looked at the effect of greater and lesser knowledge among participants about how online behavioural advertising works. Those with greater perceived knowledge were likely to see greater benefits, but also greater risk, he found. Similar to those with little perceived understanding, they tilted strongly toward privacy concerns and avoiding ads.
Ham’s study of online behavioural advertising follows from his interest in all forms of hidden persuasion, and his previous research has looked at product placement, user-generated YouTube videos and advergames. But OBA is “a very special type,” he said, in that it elicits risk perceptions and privacy concerns different from those in response to those other forms.
The study conclusions could have added significance, Ham said, because research has shown that college-age individuals, like those in his study pool, are generally less concerned about privacy than those in older age groups.
If his findings are an accurate reflection of consumer attitudes, Ham said they could represent “a really huge challenge to the advertising industry” since online behavioural advertising represents a growing segment of advertising revenue.
Ham thinks advertisers, in their own interest, may want to make the process more transparent and controllable. “They need to educate consumers, they need to clearly disclose how they track consumers’ behaviour and how they deliver more-relevant ad messages to them,” he said.
Giving consumers control is important because it might keep them open to some personalised online advertising, rather than installing tools like ad blockers, in use by almost 30 percent of online users in the U.S., he said.
With little understanding of online behavioural advertising, and no easy way to control it, “they feel a higher fear level than required, so they just block everything.”
It’s all the more important because the technology is only getting better and more accurate, Ham said. Tracking systems “can even infer where I’m supposed to visit tomorrow, where I haven’t visited yet.”
It’s easy to see why Google (and some 37,000 people) were tricked — the developer who packaged the adware into an extension used the name of an already popular and legitimate extension, AdBlock Plus.
Additionally, the bogus page in the Chrome store came with reviews. In short, the fraudulent extension looked pretty realistic. Twitter user SwiftOnSecurity, who regularly tweets about web security, posted an image of the devious extension:
Google eventually caught wind of the breach and removed the deceitful adware, but it remains unclear just how harmful the malware is for those who already downloaded the extension. At least one unfortunate user says they’re being hit with ads. In a screenshot of a review, posted by SwiftOnSecurity, the user states that the “instant this was added to Chrome started getting invasive ads with high volume levels opening new tabs.”
Though Google took down the adware, SwiftOnSecurity was unimpressed by Google’s failure to stop this malware from sneaking through and ending up conspicuously displayed in the Chrome store in the first place:
The 37,000 infected users probably hope this public shaming further motivates Google to buffer the Chrome store’s verification process. After all, malicious developers will only get more inventive if the problem isn’t fixed.
Featured Image Credit: Mark Lennihan/AP/REX/Shutterstock
Facebook has introduced Facebook Cross-Platform Brand Lift in the US and UK which along with Nielsen Total Brand Effect with Lift will help advertisers optimize their Facebook and TV campaigns using actionable results according to a blog post.
The platform will see Facebook will match rival Google which launched Brand Lift for TV some years back in order to help marketers understand how YouTube campaigns can impact metrics such as awareness.
Facebook’s advertising partners who are expanding from digital advertising into cross-media campaigns will be able to leverage Facebook Cross-Platform Brand Lift solution.
Margo Arton, senior director of Ad Effectiveness at BuzzFeed said: “Now that Buzzfeed has begun to diversify our media strategies to include both Television and Digital, having the option to leverage solutions such as Facebook’s Cross-Platform Brand Lift and Nielsen Total Brand Effect with Lift presents a great opportunity.”
“We look forward to using cross-platform brand lift measurement to both receive valuable insights about our multi-media campaign performance in a single reporting surface, and also to optimize campaign elements such as spend and creative across both platforms.”
Facebook cited an example of household brand Shark’s campaign which was deemed a success as measured by Nielsen Total Brand Effect with Lift.
Ajay Kapoor, VP, Digital Transformation & Strategy, SharkNinja said: “We proved that Facebook video ads are a natural complement to TV campaigns. We experienced better brand results among people who saw ads on both versus just TV or Facebook alone. We saw the ‘better together’ impact first-hand. Facebook and TV are powerful individually, but deliver a stronger message to our audience when used in tandem.”
Facebook and Google’s advertising platforms are out of control. That used to be a good thing. Now…not so much
Now that’s some damn precise audience targeting! From Buzzfeed.
Facebook and Google’s advertising infrastructure is one of humanity’s most marvelous creations. It’s also one of its most terrifying, because, in truth, pretty much no one really understands how it works. Not Mark Zuckerberg, not Larry Page, and certainly not Russian investigator Robert Mueller, although of the bunch, it seems Mueller is the most interested in changing that fact.
Pro Publica catches Facebook with its algorithmic pants down.
And that’s a massive problem for Facebook and Google, who have been dragged to the stocks over their algorithms’ inability to, well, act like a rational and dignified human being.
So how did the world’s most valuable and ubiquitous companies get here, and what can be done about it?
Well, let’s pull back and consider how these two tech giants execute their core business model, which of course is advertising. You might want to pour yourself an adult beverage and settle in, because by the end of this, the odds of you wanting the cold comfort of a bourbon on ice are pretty high.
In the beginning (OK, let’s just say before the year 2000), advertising was a pretty simple business. You chose your intended audience (the target), you chose your message (the creative), and then you chose your delivery vehicle (the media plan). That media plan involved identifying publications, television programs, and radio stations where your target audience was engaged.
Those media outlets lived in a world regulated by certain hard and fast rules around what constituted appropriate speech. The FCC made sure you couldn’t go full George Carlin in your creative execution, for example. The FTC made sure you couldn’t commit fraud. And the FEC — that’s the regulatory body responsible for insuring fairness and transparency in paid political speech — the FEC made sure that when audiences were targeted with creative that supports one candidate or another, those audiences could know who was behind same-said creative.
But that neat framework has been thoroughly and utterly upended on the Internet, which, as you might recall, has mostly viewed regulation as damage to be routed around.
After all, empowering three major Federal regulatory bodies dedicated to old media advertising practices seems like an awful lot of liberal overkill, n’est ce pas? What waste! And speaking of waste, honestly, if you want to “target” your audience, why bother with “media outlets” anyway?! Everyone knows that Wanamaker was right — in the offline world, half your advertising is wasted, and thanks to offline’s lack of precise targeting, no one has a clue which half that might be.
But as we consider tossing the offline baby out with the bathwater waste, it’s wise to remember a critical element of the offline model that may well save us as we begin to sort through the mess we’re currently in. That element can be understood via a single word: Context. But we’ll get to that in a minute. First, let’s go back to our story of how advertising has shifted in an online world, and the unintended consequences of that shift (if you want a even more thorough take, head over to Rick Webb’s NewCo Shift series: Which Half Is Wasted).
Google: Millions Flock to Self Service, Rise of the Algos
Back in the year 2000, Google rolled out AdWords, a fantastically precise targeting technology that allowed just about anyone to target their advertisements to…just about anyone, as long as that person was typing a search term into Google’s rapidly growing service. (Keep that “anyone” word in mind, it’ll come back to haunt us later.) AdWords worked best when you used it directly on Google’s site — because your ad came up as a search result right next to the “organic” results. If your ad was contextually relevant to a user’s search query, it had a good chance of “winning” — and the prize was a potential customer clicking over to your “landing page.” What you did with them then was your business, not Google’s.
As you can tell from my fetishistic italicization, in this early portion of the digital ad revolution, context still mattered. Google next rolled out “AdSense,” which placed AdWords on publishers’ pages around the Internet. AdSense didn’t work as well as AdWords on Google’s own site, but it still worked pretty well, because it was driven by context — the AdSense system scanned the web pages on which its ads were placed, and attempted to place relevant AdWords in context there. Sometimes it did so clumsily, sometimes it did so with spectacular precision. Net net, it did it well enough to start a revolution.
Within a few years, AdWords and AdSense brought billions of dollars of revenue to Google, and it reshaped the habits of millions of advertisers large and small. In fact, AdWords brought an entirely new class of advertiser into the fold — small time business owners who could compete on a level playing field with massive brands. It also reshaped the efforts of thousands of publishers, many of whom dedicated small armies of humans to game AdWords’ algorithms and fraudulently drink the advertisers’ milk shakes. Google fought back, employing thousands of engineers to ward off spam, fraud, and bad actors.
AdWords didn’t let advertisers target individuals based on their deeply personal information, at least not in its first decade or so of existence. Instead, you targeted based on the expressed intention of individuals — either their search query (if on Google’s own site), or the context of what they were reading on sites all over the web. And over time, Google developed what seemed like insanely smart algorithms which helped advertisers find their audiences, deliver their messaging, and optimize their results.
The government mostly stayed out of Google’s way during this period.
When Google went public in 2004, it was estimated that between 15 to 25 percent of advertising on its platform was fraudulent. But advertisers didn’t care — after all, that’s a lot less waste than over in Wanamaker land, right? Google’s IPO was, for a period of time, the most successful offering in the history of tech.
Facebook: People Based Marketing FTW
Then along came Facebook. Facebook was a social network where legions of users voluntarily offered personally identifying information in exchange for the right to poke each other, like each other, and share their baby pictures with each other.
Facebook’s founders knew their future lay in connecting that trove of user data to a massive ad platform. In 2008, they hired Sheryl Sandberg, who ran Google’s advertising operation, and within a few years, Facebook had built the foundation of what is now the most ruthlessly precise targeting engine on the planet.
Facebook took nearly all the world-beating characteristics of Google’s AdWords and added the crack cocaine of personal data. Its self service platform, which opened for business a year or so after Sandberg joined, was hailed as ‘ridiculously easy to use.’ Facebook began to grow by leaps and bounds. Not only did everyone in the industrialized world get a Facebook account, every advertiser in the industrialized world got themselves a Facebook advertising account. Google had already plowed the field, after all. All Facebook had to do was add the informational seed.
Both Google and Facebook’s systems were essentially open — as we established earlier, just about anyone could sign up and start buying algorithmically generated ads targeted to infinite numbers of “audiences.” By 2013 or so, Google had gotten into the personalization game, albeit most folks would admit it wasn’t nearly as good as Facebook’s, but still, way better than the offline world.
So how does Facebook’s ad system work? Well, just like Google, it’s accessed through a self-service platform that lets you target your audiences using Facebook data. And because Facebook knows an awful lot about its users, you can target those users with astounding precision. You want women, 30–34, with two kids who live in the suburbs? Piece of cake. Men, 18–21 with an interest in acid house music, cosplay, and scientology? Done! And just like Google, Facebook employed legions of algorithms which helped advertisers find their audiences, deliver their messaging, and optimize their results. A massive ecosystem of advertisers flocked to Facebook’s new platform, lured by what appeared to be the Holy Grail of their customer acquisition dreams: People Based Marketing!
The government mostly stayed out of Facebook’s way during this period.
When Facebook went public in 2012, it estimated that only 1.5% of its nearly one billion accounts were fraudulent. A handful of advertisers begged to differ, but they were probably just using the system wrong. Sad!
Facebook’s IPO quickly became the most successful IPO in the history of tech. (Till Alibaba, of course. But that’s another story).
(Meanwhile, Programmatic.)
The programmatic Lumascape. Seems uncomplicated, right?
Stunned by the rise of the Google/Facebook duopoly, the tech industry responded with an open web answer: Programmatic advertising. Using cookies, mobile IDs, and tons of related data gathered from users as they surfed the web, hundreds of startups built an open-source version of Facebook and Google’s walled gardens. Programmatic was driven almost entirely by the concept of “audience buying” — the purchase of a specific audience segment regardless of the context in which that audience resided. The programmatic industry quickly scaled to billions of dollars — advertisers loved its price tag (open web ads were far cheaper), and its seemingly amazing return on investment (driven in large part by fraud and bad KPIs, but that’s yet another post).
Facebook and Google were unfazed by the rise of programmatic. In fact, they bought the best companies in the field, and incorporated their technologies into their ever advancing platforms.
The Storm Clouds Gather
But a funny thing happened as Google, Facebook and the programmatic industry rewrote advertising history. Now that advertisers could precisely identify and target audiences on Facebook, Google and across the web, they no longer needed to use media outlets as a proxy for those audiences. Media companies began to fall out of favor with advertisers and subsequently fail in large numbers. Google and Facebook became advertisers’ primary audience acquisition machines. Marketers poured the majority of their budgets into the duopoly — 70–85% of all digital advertising dollars go to the one or the other of them, and nearly all growth in digital marketing spend is attributable to them as well.
By 2011, regulators began to wrap their heads around this burgeoning field. Up till then, Internet ads were exempt from political regulations governing television, print, and other non digital outlets. In fact, both Facebook and Google have both lobbied the FEC, at various times over the past decade or so, to exclude their platforms from the vagaries of regulatory oversight based on an exemption for, and I am not making this up, “bumper stickers, pins, buttons, pens and similar small items” where posting a disclaimer is impracticable (sky writing is also mentioned). AdWords and mobile feed ads were small, after all. And everyone knows the Internet has limited space for disclaimers, right?
Anyway, that was the state of play up until 2011, when Facebook submitted a request to the FEC to clear the issue up once and for all. With a huge election coming in 2012, it was both wise and proactive of Facebook to want to clarify the matter, lest they find themselves on the wrong end of a regulatory ruling with hundreds of millions of dollars on the line. (Of course, they favored exemption over actual regulation).
The FEC failed to clarify its position, but did request comment from industry and the public on the issue (PDF). In essence, things remained status quo, and nothing happened for several years.
That set the table for the election of 2016. In October of that year, perhaps realizing it had done nothing for half a decade while the most powerful advertising machine in the history of ever slowly marched toward its seemingly inevitable date with emergent super intelligence, the FEC re-opened its request for comments on the whether or not political advertising on the Internet should have some trace of transparency. But that was far too late for the 2016 election.
Most everyone I speak to tells me that last week’s revelations about Facebook, Russia, and political advertising is, in the words of Senator Mark Warner, “the tip of the iceberg.” Whether or not that’s true (and I for one am quite certain it is), it’s plenty enough to bring the issue directly to the forefront of our political and regulatory debate.
Now the news is coming fast and furious: At what was supposed to be a relatively quotidian regular meeting of the FEC this week, the commissioners voted unanimously to re-open (again) the comment period on Internet transparency. The Campaign Legal Center, launched in 2002 by a Republican ally of Senator John McCain (co-sponsor of the McCain Feingold Bipartisan Campaign Reform Act of 2002), this week issued a release calling for Facebook to disclose any and all ads purchased by foreign agents. (Would that it were that simple, but we’ll get to that in the next installment.) One of the six FEC commissioners, a Democrat, subsequently penned an impassioned Op Ed in the Washington Post, calling for a new regulatory framework that would protect American democracy from foreign meddling. The catch? The Republicans on the commission refuse to consider any regulations unless the commission receives “enough substantive written comments.”
Once the link for comments goes up in a week or two, I’m pretty sure they will.
But in the meantime, there’s plenty of chin stroking to be done over this issue. While this may seem like a dust up limited to the transparency of political advertising on the internet, the real story is vastly larger and more complicated. The wheels of western capitalism are greased by paid speech, and online, much of that speech is protected by the first amendment to our constitution, as well as established policies enshrined in contract law between Facebook, Google, and their clients. There are innumerable scenarios where a company or organization demands opacity around its advertising efforts. So many, in fact, that if I were to go into them now, I’d extend this piece by another 2,500 words.
And given I’m now close to 3,000 words in what was supposed to be a 600-word column, I’m going to leave exploring those scenarios, and their impact, to next week’s columns. In the meantime, I’ll be speaking with as many experts and policy folks from tech, Washington, and media as I can find. Suffice to say, big regulation is coming for big tech. Never in the history of the tech industry has the 1996 CDMA ruling granting tech platforms immunity from the consequences of speech on their own platforms been more germane. Whether it’s in jeopardy or not remains to be seen.
This is not a simple issue, and resolving it will require a level of rational discourse and debate that’s been starkly absent from our national dialog these past few years. At stake is not only the fundamental advertising models that built our most valuable tech companies, but also the essential forces and presumptions driving our system of democratic capitalism*. Not to mention the nascent but utterly critical debate around the role of algorithms in civil society. And as we explore solutions to what increasingly feels like an intractable set of questions, we’d do well to keep one word in mind: Context.
Over the last few days, a slew of reporting, inspired by ProPublica, has revealed that it’s actually quite easy, through the programmatic structure of most online advertising, to create ads meant to target those who have espoused racist, antisemitic, or other hateful ideas.
Here’s a quick rundown of the major internet companies, and what has been discovered about their advertising platforms:
Facebook
On Sept. 14, ProPublica reported that Facebook allowed advertisers to target categories and ideas such as “Jew hater,” “How to burn jews,” and “History of ‘why jews ruin the world,’” based on interest Facebook users had expressed on the social network and terms with which they had used to describe themselves.
While Facebook removed those categories after ProPublica’s investigation, Slate then discovered that there are dozens of other racist, sexist, and xenophobic categories which advertisers could potentially target. It took Facebook less than a minute to approve ads against phrases like “Kill Muslimic Radicals”and “Ku-Klux-Klan,” and Slate found myriad other options, like “Killing Bitches,” “Killing Hajis,” and “Nazi Party (Canada).”
Facebook released a statement yesterday after ProPublica’s report, saying in part:
Keeping our community safe is critical to our mission. And to help ensure that targeting is not used for discriminatory purposes, we are removing these self-reported targeting fields until we have the right processes in place to help prevent this issue. We want Facebook to be a safe place for people and businesses, and we’ll continue to do everything we can to keep hate off Facebook.
Google
BuzzFeed discovered similar targeting issues on Google’s AdWords platform, which runs the advertisements you see on Google search results pages. Typing in keyword suggestions (which advertisers use to build their ads and figure out who to target) like “why do jews ruin everything” led to the system generating more keyword suggestions like “jews ruin the world” and “jewish parasites.” Buzzfeed was also able to build and launch a campaign around the phrase “black people ruin neighborhoods.”
When Quartz attempted to recreate BuzzFeed’s efforts using similar terms, or terms like those used by ProPublica and Slate, no keyword suggestions were returned. Google has since disabled many of the keywords that BuzzFeed tested.
Sridhar Ramaswamy, Google’s senior vice president in charge of ads, told Quartz in a statement:
Our goal is to prevent our keyword suggestions tool from making offensive suggestions, and to stop any offensive ads appearing. We have language that informs advertisers when their ads are offensive and therefore rejected. In this instance, ads didn’t run against the vast majority of these keywords, but we didn’t catch all these offensive suggestions. That’s not good enough and we’re not making excuses. We’ve already turned off these suggestions, and any ads that made it through, and will work harder to stop this from happening again.
Twitter
The Daily Beast was able to target similarly derogatory demographics on Twitter. It reported:
Twitter’s advertising platform tells prospective marketers it has 26.3 million users interested in the derogatory term “wetback,” 18.6 million accounts that are likely to engage with the word “Nazi,” and 14.5 million users who might be drawn to “n**ger.”
A Twitter representative told Quartz about the Daily Beast’s report:
The terms cited in this story have been blacklisted for several years and we are looking into why the campaign cited in this story were able to run for a very short period of time. Twitter actively prohibits and prevents any offensive ads from appearing on our platform, and we are committed to understanding 1) why this happened, and 2) how to keep it from happening again.
Snapchat
Quartz checked on Snapchat’s advertising platform to see if we were able to target using similar terms used on the other platforms. We were not able to: It seems that Snapchat’s demography isn’t quite as granular as the other platforms, which are far more text-based than Snapchat, and so it’s likely easier for them to glean what sorts of things its users are sharing than through all the videos and images posted to Snapchat.
Bing
Microsoft’s second-placed search network seems to have a similar problem to its other platforms. When Quartz created a test advertising campaign on Bing Ads, we weren’t able to directly target specifically loaded terms, but searching for just about any phrase in Bing’s “keyword suggestions” generator will generate specific keywords that you might want to try to target instead. Here’s one example, using “Hitler” as the search term:
(Screenshot/Bing Ads)
A representative for Bing told Quartz:
We take steps to ensure our Bing Ads always meet reasonable standards. We are committed to working with partners who share our vision for relevant, impactful brand interaction and respect the integrity of consumer choice.
Yahoo
Quartz attempted to create an ad campaign on Yahoo, but it seems there’s no simple way to create one online without speaking to a representative from Oath (Yahoo’s parent company) first. And presumably fewer people would feel comfortable telling a sales rep the sorts of things they’re targeting than they would inputting them into a computer system. Hopefully.
LinkedIn
Microsoft’s professional social network doesn’t seem to let users target based on arbitrary phrases or demographics. Other than geography, these are the only things you can target against on LinkedIn:
(Screenshot/LinkedIn)
The only section that might have the potential for hateful terms would be in “Member groups”—but a cursory search of terms like those used above didn’t reveal many professional hate groups to target on the platform. We did, however, come across this group:
(Screenshot/LinkedIn)
Upon further inspection, however, it seems that this group was set up by a LinkedIn employee trying to see whether they could set up a group with a title like this. Obviously, it worked:
(Screenshot/LinkedIn)
LinkedIn sent Quartz the following statement:
Hate has no place on LinkedIn and will not be tolerated. When we are made aware of such content, we act swiftly to enforce our policy and remove said content. On Friday, a member of our team created a group solely for internal testing purposes and after a brief testing period, we took the group down.
They’re called the “duopoly” of online advertising. Facebook and Google account for 75% of the U.S. digital ad spend — and almost all of its growth, according to the Interactive Advertising Bureau.
Facebook reported 45% growth in the last quarter and Google’s parent company Alphabet posted earnings of $26 billion, 87% coming from advertising revenue.
But are these behemoths about to be blindsided by a fierce competitor with a better ROI?
A consumer research study for a beverage manufacturer uncovered an interesting trend, one that might tip the scale for advertisers. Consumers who had an Amazon Prime account started their search for a purchase at Amazon 100% of the time. If they knew what they wanted to buy, they went directly to Amazon to search for different brands with the best price and delivery options.
With 85 million Amazon Prime members as of June 2017, it’s not going to take long for consumer brands to discover that if you want to invest ad dollars in finding buyers with high purchase intent and conversion rates, Amazon is going to be hard to ignore.
Although its ad business is small in comparison to Google and Facebook, with only 1% of global ads, it actually is one of Amazon’s fastest growing areas, now on track to generate close to $2 billion this year.
Amazon also offers organizations a broad spectrum of advertising products, ranging from its ad platform offering mobile and desktop display and banner ads, to dynamic and coupon ads. Customer campaign pages allow advertisers to create immersive cross-platform landing pages that can display more than one product.
With the digital ad market predicted to grow at 16% this year to $83 billion according to eMarketer, the Facebook-Google duopoly will get its fair share, and almost all of the attention, especially considering the growth of Facebook’s Snapchat ad revenue, up 158% in the past year. And that may be just how Amazon likes it. It has a history of sneaking up on competitors. Just ask Microsoft and IBM about Amazon Web Services (AWS).
Andy Jassy, the AWS CEO, said that in some ways the growth of his business was a classic case of disruption dynamics. “The competition simply didn’t believe there was enough of a market to worry about it. The dominant players don’t have any reason to worry about someone attacking the bottom of the market.” AWS now owns a third of the cloud infrastructure services market, more than three times that of its next closest competitors.
Amazon seems to follow Al Pacino’s “never let them see you coming” advice from “The Devil’s Advocate.” But one ad executive — Sir Martin Sorrell, WPP’s CEO — has noticed. In a recent interview with Bloomberg said, Sorrell said, “The company that would worry me if I was a client — or I think worries our clients, more than Google and Facebook — is Amazon.”
Smart ad dollars follow consumer behavior — and, as we just learned, those consumers are headed to Amazon.
Google has announced its AdWords ads will undergo one of the most significant makeovers in recent memory later this month.
The forthcoming updates will apply to mobile ads only, and will change how sitelinks, callouts, and structured snippets are displayed.
Sitelinks Carousel
Google will be introducing a sitelinks carousel to AdWords ads, which resembles the sitelinks that currently appear for organic listings.
“Going forward, we’re simplifying how mobile sitelinks will show by using both horizontal buttons and larger vertical links.”
The company says searchers are two times as likely to interact with sitelinks when they appear in this new format.
Callouts and Snippets
When this AdWords update rolls out, callouts and structured snippets will blend in with the ad copy. Instead of appearing in separate lines below the ad, they will appear in paragraph form.
With this change to callouts and structured snippets, more of them will be eligible to show up with AdWords ads going forward.
Google says user studies have found this new format to be “more informative and engaging.”
Changes on the Advertisers’ End
It doesn’t sound like there will be any changes on the end of the advertiser. The new formatting will take place on Google’s end, while advertisers can continue to run their campaigns as they usually would.
