Democracy & Government

Our Next President: Also Brought to You by Big Data and Digital Advertising

How the Trump campaign used big data to elect a president.

Big Brother Brings You President Donald Trump

Facebook Election Day badge (Photo by Joey Kotfica/Contributor/Getty Images)

The ease of promoting and profiting from fake news, the use of social media to spread disinformation and twist the truth and the ability to microtarget voters on all their devices reveal how political campaigns and special-interest groups are taking full advantage of the “Big Data”-driven digital marketing system that Google and Facebook have helped unleash. Campaigns now have needle-in-the-haystack capabilities, provided by commercial marketing and media companies, to find and motivate an individual voter. The Trump team, for example, was able to identify someone from rural America who felt “disenfranchised,” had not voted in recent elections (and didn’t show up in polls as a likely voter), and then use a data-and-digital-concocted brew to trigger emotions and behavior.

Behind the digital curtain that informs and entertains is a commercial surveillance system whose function until recently has been primarily to sell us junk food, credit cards and retail products.

Behind the digital curtain that informs and entertains is a commercial surveillance system whose function until recently has been primarily to sell us junk food, credit cards and retail products. But in 2016, the system was fully deployed for political gain. Just take a look at the techniques used by the Trump, Clinton, Sanders and other political campaigns that describe how they reached and influenced voters. They are right out of the digital advertisers playbook: IP address targeting, geofencing, pixel profiles, Data Management Platforms (DMPs), algorithms, data onboarding, moment scoring, header bidding, and programmatic advertising. Electoral campaigns — from all sides — are fusing cookies (data files placed on our web browsers) with information gathered by consumer data companies along with our voter files, which illustrates how our commercial pursuits and political interests are being merged today.

Advances in the best way to use all the information routinely captured and analyzed about us, and which has been greatly expanded by our use of mobile phones and apps, have helped companies to create highly detailed personal dossiers that merge our online and offline information, make predictions about how we — or others like us — will respond, and then orchestrate desired behaviors. Google boasts to its advertisers, for example, that it can help them connect with us in a “micromoment” — just at the point when we may make some decision. Facebook, for its part, assures marketers they can use its system to influence the decisions of “real people.”

Take A Look: No Escaping Dragnet Nation

There is little standing in the way of a Trump, Clinton or any other political effort from easily acquiring access to the huge reams of personal information now available through commercial data companies, sold by publishers and online enterprises and used to target us when we log on to social media such as Facebook. The digital marketing and data industry has largely been immune from any regulation under the Obama administration, either on consumer data collection practices or business operations. The US is still one of the only advanced-economy countries without comprehensive privacy legislation, and what little we have is already under threat. For the vast majority of Americans, there is practically no legal way to turn off or stem the flow of their information. The government’s “hands-off” approach to how Facebook, Google and the rest of the digital marketing industry operate has helped unleash powerful and unaccountable forces that helped elect our next president.

Political campaigns can buy data from internet and other companies about our finances, health concerns, race, ethnicity, shopping behavior and geo-location, along with what we read online and what our political interests are. Leading data companies including Nielsen, Neustar, Oracle and Acxiom sell this information to political campaigns and others working to influence public opinion. Political data firms such as Cambridge Analytica, which worked for Trump, layer on predictive modeling and psychological tools to help identify “personality traits,” behavior and “what really drives … decision making.” Trump’s White House chief strategist Steve Bannon sits on its board. This company, which also helped support the Brexit “Leave” initiative, says it has compiled “5,000 data points on over 220 million Americans.” Cambridge Analytica is the US affiliate of the SCL Group in the UK, which engages in political, commercial and defense work (including “PSYOPS” — psychological operations). According to its website, SCL’s specialties include “behavioural polling, behavioural microtargeting, StratCom and target audience analysis” (a service it is helping NATO perfect).

SCL’s operations illustrate how data analysis is being used today, from trying to win “hearts and minds” in Afghanistan to selling us Big Macs or a President Donald Trump. According to the company, they “combine commercial and public big-data sets with large-scale quantitative research to predict everything from whether people are likely to vote through to what products and services they are most likely to buy.” Its work with political campaigns takes advantage of the “predictive power of data” and uses “physio-lingual analysis tools to uncover subconscious … reasoning.”

The Trump campaign also used digital targeting to conduct “major voter suppression operations,” according to a report in Bloomberg, focusing on African-Americans, young women and white liberals.

Such capabilities benefited Trump, who only won because, as Cambridge Analytica itself wrote, “The president-elect flipped Florida, Pennsylvania, Michigan and Wisconsin by margins of less than 2 percent. If Hillary Clinton had taken those last three states, she would have won the election. Trump won those three states by a combined margin of approximately 107,000 votes.” The company also explained:

The 2016 presidential election showed that the use of data to identify, persuade and turn out voters has become increasingly sophisticated. Cambridge Analytica’s data science, digital marketing and research teams informed key decisions on campaign travel, communications and resource allocation…. Every week Cambridge Analytica collected responses from 1,500 to 2,000 people in each battleground state. It used this research and data to model scores for all voters across key states: which candidate they preferred, which were “persuadable,” the issues they cared about and how likely they were to actually vote on Election Day. Every voter in each battleground state was also segmented by ethnicity, religion and the issues that concerned them most….

When, in the final weeks of the race, the firm’s data scientists recalculated voter turnout and recalibrated their models to show how Donald Trump could win, the GOP candidate revisited states like Michigan and Wisconsin…. Online ads placed by the firm were viewed a staggering 1.5 billion times by millions of Americans ….

If Hillary Clinton’s campaign was the Titanic, then Donald Trump’s campaign was a speedboat: nimble, flexible and able to adapt fast.”

The Trump campaign also used digital targeting to conduct “major voter suppression operations,” according to a report in Bloomberg, focusing on African-Americans, young women and white liberals — with tools including an onslaught of what are called “dark posts” on Facebook. These targeted messages included “a South Park-style animation of Hilary Clinton delivering her infamous 1996 ‘super predator’ remarks.” Marketers use dark posts to deliver targeted content to only on the Facebook newsfeeds of individuals, not on the ones operated by the companies that generate them. Giles-Parscale, another online marketing firm working for Trump, was part of “Project Alamo,” a digital data initiative spearheaded by Trump son-in-law Jared Kushner. “Facebook and Twitter were the reason we won this thing,” said Brad Parscale, who is now assembling a nonprofit advocacy group designed to press Trump’s agenda from outside the White House. How Trump, Kushner and their operatives use the considerable data assets controlled by the White House (which was expanded under Obama) and work with their new outside digital operation should raise concerns, at the very least, from civil libertarians and others.

Google and Facebook likely will continue to be friendly ports of call for Trump and the GOP. They provide a host of services to both major political parties’ campaigns and have benefited from the Big Money pouring into them to target voters online. To help campaigns “understand the influence of digital media and online video in the 2016 elections,” Google hired Republican pollster Frank Luntz, along with Julie Hootkin from the Global Strategy Group, to offer insights as “guest editors” on the company’s principal website set up for advertisers. Google also promoted ways political advertisers could take advantage of YouTube, its search engine and all its other tools for marketers. It offered this advice: “Voter decisions used to be made in living rooms, in front of televisions. Today, they’re increasingly made in micromoments, on mobile devices.” The GOP was reported to have awarded Google a major share of its “largest digital ad deal ever,” earmarking $150 million to buy online video ads, according to Ad Age last May.

judicialwatch

Facebook also offers targeting opportunities for political campaigns, which are encouraged to buy ads and other products to appeal to its more than 162 million US users, enabling campaigns to target by age, gender, congressional district and interests. Campaigns can also supply their own data to Facebook, which will help them target individuals. Facebook has promoted as an advertising “success” story its fundraising targeting for the conservative group Judicial Watch, which included a mobile ad (pictured right) prominently featuring a headshot of Hillary Clinton alongside the headline, “Hillary’s Email Scandal Exposed.” The case study explains, “Advanced matching for Custom Audiences is a targeting tool that allows advertisers to upload multiple data points at once to create an audience on Facebook. Data points include general information like first and last name, ZIP code, state, country, age and gender, and also email, phone number, mobile advertiser ID and Facebook app user ID.”

So many companies now specialize in or supply data to political campaigns, as well as offer ways to influence voters more precisely. There are even data made available to Republican campaigns by Democratic-oriented companies. For example, data analytics specialist L2, which has been helping Donald Trump since early 2015, offers a predictive modeling product that scores voters on “likelihood of support” on dozens of issues, including their views of Black Lives Matter, church attendance, fracking, gun laws, gay marriage, the Mexican border wall, transgender bathrooms and many more. HaystaqDNA, a political data firm started by operatives of President Obama’s previous campaigns, sold some of its own data to L2 (and which were used, according to a former Trump data official, to “pinpoint a specific type of would-be supporter.”)

But it’s more than just data gathering and online targeting that concerns privacy and democracy advocates. Pursuing monetization at any cost overwhelms what’s left of the boundaries that separate news from advertising or purposefully deceptive content — and the public is being conditioned to accept fake as real. The online giants have effectively fashioned a brand-focused online echo chamber in which the public is continually encouraged to accept, and virally pass along, advertising content that is purposefully disguised as legitimate news and information. That chamber is also home to a growing effort by leading advertisers to use the latest neuromarketing techniques to subconsciously imprint on people a more favorable view of their brands.

While political campaigns will argue the First Amendment gives them the right to buy such data, we shouldn’t allow retailers, data brokers, online marketers and publishers to gather and use our personal information in the first place. Of course, Trump and the GOP-dominated Congress are not likely to pass any new privacy laws or impose regulations on how commercial data can be gathered and analyzed. Already, the GOP is said to be ready to reverse the Federal Communication Commission’s privacy rules that were adopted last October. The Federal Trade Commission is likely to become even weaker on data issues and governing consumer data marketing practices.

While we focus on the need to protect the regulations we have, and demand even more, we must also call for greater responsibility from Google, Facebook and the digital industry — including leading publishers of news and information — about their data and marketing practices. It’s not only in the UK and US where these tactics are playing a destructive role in democracy; it’s also been exported to Germany, Italy and other countries. There is urgency to do so, as the industry is moving into place the next generation of personalized micropersuasion practices that harness artificial intelligence, cognitive computing, virtual reality and ubiquitous data collection that will far surpass what is being used today.

Jeff Chester

Jeff Chester is executive director of the Center for Digital Democracy, a Washington, DC, nonprofit group that helps hold accountable the digital media industry. Follow the CDD on Twitter: @digitaldemoc.

RELATED CONTENT