There was a sort of symbolism in the way President Donald Trump and his inner circle formulated and rolled out last Friday’s executive order blocking entry to the United States from seven Muslim-majority nations; according to Politico:
Senior staffers on the House Judiciary Committee helped Donald Trump’s top aides draft the executive order curbing immigration from seven Muslim-majority nations, but the Republican committee chairman and party leadership were not informed, according to multiple sources involved in the process…
It’s extremely rare for administration officials to circumvent Republican leadership and work directly with congressional committee aides…GOP leaders received no advance warning or briefings from the White House or Judiciary staff on what the executive order would do or how it would be implemented…Republicans on the Hill spent the entire weekend scrambling to find out what was going on, who was involved and how it was that they were caught so flat-footed.
“Caught so flat-footed” is basically the entire story of Trump and the Republican establishment (and later, Hillary Clinton and the Democrats), and what makes this a story for Stratechery is that, as with so many other things in our world, technology is directly implicated.
Political Parties and the Media
There is little question that social media is the most important factor in Trump ascending to the presidency. No, not because of fake news — new research confirms the common sense conclusion that made-up stories almost exclusively appeal to people who have already made up their minds — but rather the dilution of “real” news, or more broadly, the media as a whole.
As I explained last year, in a world where most voters only had access to one or two newspapers, or three or four TV stations, the sheer logistics of gaining the sort of nationwide awareness necessary for a viable presidential campaign required institutional support — political parties, specifically. This was the core thesis of The Party Decides, a conventional wisdom-defining book which held that, even though the presidential candidate selection process had shifted from smoke-filled rooms to democratic primaries (and somewhat less democratic caucuses), in practice political parties (specifically, the activists who actually cared about outcomes and thus did the work and raised the money) still had veto power on their candidates.
What so many missed, though, was that this definition of political parties and the roles they play was inextricably tied up with the media — specifically, the pre-Internet media. As long as there were only two ways to reach voters — “paid” media (i.e. advertisements) and “earned” media (i.e. news coverage) — then those who raised the money and made the news had the power to end most campaigns before they could even begin.
However, as has been well-documented on this site and many others, the media industry has, thanks to the Internet, been completely stripped of its gatekeeper role when it comes to the spread of information. Instead of scarce newspapers or TV stations there is an abundance of information providers, which means the real power has shifted from distribution to discovery.
Thus, by extension, the real power in politics has shifted from parties to the people.
The first company to benefit from the shift to abundance was Google, which reduced newspapers to articles and proceeded to give you exactly what you were looking for; in 2008 the search engine’s increasing importance paid off in a big way for a then relatively-unknown Senator from Illinois named Barack Obama, who both dominated Google Trends and invested disproportionately in Internet advertising generally and search advertising in particular.
Eight years later the dominant force in discovery is Facebook; whereas Google gave answers, Facebook doesn’t even require you to ask a question. And, once again, the winning candidate was the one who dominated the new metrics: Hillary Clinton may have had 500 newspapers and magazine endorsements to Trump’s 27, and may have spent more than twice as much as the Republican nominee on television ads, but the ratio was reversed when it came to digital advertising, and perhaps most tellingly Trump crushed Clinton when it came to Facebook activity, with over 960 million likes, comments, shares & posts, as compared to Clinton’s 410 million.
Interestingly, there was one candidate who rivaled Trump on Facebook; according to FiveThirtyEight, as of April 18 Bernie Sanders had 25% of all Presidential ‘likes’, exceeding Trump’s 24%:
There are obviously caveats to this data, as FiveThirtyEight notes:
Be careful how you interpret these numbers: Facebook likes are not votes. According to the Pew Research Center, 58 percent of American adults use Facebook. But this share is not a representative sample of the country — Facebook users are disproportionately young (although not as young as users of other social media networks), low-income and female. And the sample may be even more skewed because only some people on Facebook have liked a presidential candidate’s page and because those pages haven’t existed for the same amount of time. As “The Literary Digest” taught us in 1936, large but biased samples aren’t so effective.
I’ll add my own caveat: this isn’t a political blog, and I’m making no judgment on how Sanders may have fared in a hypothetical face-off with Trump. Rather, there is a broader takeaway about the Internet’s impact.
Breaking Through on the Internet
Consider the mechanics of reaching voters/customers/users:
- Before the Internet, when distribution was the bottleneck, the optimal strategy was to maximize the available throughput. The best example is consumer packaged goods: companies like Proctor & Gamble built massive brands that were designed to appeal to the broadest swaths of population possible, maximizing the return on the effort and expense necessary to advertise and secure retail space. In the case of politics, this manifested as a push by the parties for broadly acceptable candidates who could appeal to the middle.
- Internet companies, on the other hand, have effectively infinite throughput. Amazon, for example, unbound by the need for shelf space and capitalizing on its transformation into an e-commerce platform, can plausibly bill itself as “The Everything Store”; products are found not through browsing but by search. This, by extension, means that products need to be wanted, not simply recognized — and the same goes for Google’s impact on politics.
- Facebook, as is its wont, supercharges these effects: instead of users “pulling” out content they are interested in, the algorithm “pushes” content based on its capability of driving engagement. And what drives engagement? Emotion and passion. That may mean a funny product video, or, in the case of politics, politicians who eschew the middle and run to the extremes.
Given the fundamentally different mechanics of Internet distribution, those Facebook numbers make a lot more sense: the extremes inspire passion which drives engagement; “broadly acceptable” doesn’t go anywhere.
This has profound implications for products and politics. First and foremost, it is fundamentally misguided to simply view “digital” as another channel that you layer on top of traditional marketing/campaign tactics like TV advertisements. In fact, products and politicians designed for the TV age — that is, meant to be palatable to the greatest number of people — are at a fundamental disadvantage on platforms like Facebook. The products and politicians that win inspire passion, stirring up a level of engagement that breaks through on a scale that far exceeds an ad buy. To put it another way, above I mentioned “paid” media and “earned” media; what matters on the Internet is “inspired” media.
The second implication is just as profound: campaigns — both for products and presidential candidates — used to be discrete events. This too sprang from the constraints of media: it takes a significant logistical effort to get a campaign off the ground. That, however, is not the case for “inspired” media: customers/voters are not passive recipients, they are active participants, and the speed with which a campaign can be created is breathtaking.
Consider the protests that erupted in response to that executive order: in a matter of hours tens of thousands of people were marching at airports around the country, driven not by professional politicians running a campaign but in response to exhortations on social media — and, as with any campaign, there was a lot of money raised as well.
The broader takeaway is that the Internet is the Rubicon: products, politicians, and strategies that were optimal on one side are suboptimal on the other. There is little to be gained from “layering on” a digital strategy to a broadly acceptable mass market offering; to succeed on the Internet the pursuit of passion must be the goal from the beginning.