7 Questions with Social Media Strategy Expert Dr. Alan Rosenblatt

Blue background with pythagorean symbol, lightbulb, big red question mark

Social Media Strategy is Imperative in the Age of Digital Politics

Dr. Alan Rosenblatt is an expert in social media strategy. He is the Director of Digital Research for Lake Research Partners, a Partner and Lead Digital & Social Media Strategist at turner4D, and a Principal at Unfiltered.Media with 30 years of experience in the digital politics industry. He specializes in digital communications at the nexus of politics, advocacy, media, and education. Dr. Rosenblatt currently serves as an adjunct professor at the George Washington Graduate School of Political Management, Johns Hopkins University, American University, and Baruch College. We talked to Alan about social media strategy and his extensive experience in digital media and politics to get his insights: 

1. How did you end up in digital politics? How have you seen the industry change?

I started using computers my senior year of college (that was a year before the Mac came out). When I arrived at American University three years later in 1986 to study for my Ph.D. in Political Science, I was assigned to a research methods professor who marched me into the Social Science Research Lab (a computer lab) and told me I had two weeks to learn how to program SPSS on the mainframe before her students would come in for help. I spent a lot of time in that lab (20 hours a week) and had a lot of down time. So I learned a lot about many kinds of computers and apps – 2 mainframe operating systems, 4 statistical software packages, SuperMacs, Windows, and a Novell Network. I worked in the American University computer labs in various capacities for the next six years.

At the same time, I was using Pagemaker to desktop-publish my own magazine (with a friend, Ben Barnett), and I was writing my dissertation on how US Presidents use television to influence public opinion. 

I was also part of the team that set American up as an internet node in 1988. I was the “end user guy.” I was not a technologist; I am all about how people use the technology. My job was to stress the system to try and break it (then report what I did to break it to the technologists) and to add all of the internet features to the mainframe menu system. That, along with my Pagemaker work, word processing with Waterloo Script on the mainframe, and tinkering with early versions of Apple’s Hypercard software taught me all about information architecture and gave me the skills that served me well when building websites for my early clients.

I also had enough background in analytical philosophy and programming in SPSS (and later in SAS) that I could figure out what could be programmed and communicate it to someone who could code. That gave me the ability to effectively evaluate advocacy software and persuasively encourage the addition of features to many of the products that other people were building.

Then, in the spring of 1993, my dissertation completed, my roommate showed me Mosaic, the first web-browser. As I clicked on the links trying to figure out what this was, all I could think was how this was going to change politics forever. 

By that time, I was a full-time political science professor at George Mason University. I had the luxury of picking a research topic and running with it. After a couple years of research, I convinced the chair of my department to allow me to teach The Politics of Cyberspace: The Luddite Versus the Technofreak. It was the first course on digital politics offered anywhere in the world (as far as I have been able to determine). Michael Cornfield, at GWU, had a class session in his political communication course the previous year. And Robert Dudley co-taught the course with me at GMU for the first two years of the six that it was offered.

Over the next six years, while teaching the course, I started providing digital and social media strategy advice (informally) to a few organizations and campaigns, including Rock the Vote and Pat Paulson’s 1996 Presidential campaign. Then I co-founded (again with Ben Barnett) a streaming video company in Philadelphia called Media Bureau Networks. MBN was the first company to drop a high-speed internet line into Northern Liberties, the Philly neighborhood that is now the city’s tech hub. I also spearheaded MBN’s acquisition of press credentials for the 2000 Presidential nominating conventions in Philly and LA, where we streamed four talk shows a day (two of which I created) and interviewed Steve Forbes, Jr., the Reverend Jerry Falwell, Sr., Rep. Bob Barr, George P. Bush, Ted Nugent, and Schooly D, among others, all live and online. 

In 2000 I left GMU. My work in digital politics and social media strategy, groundbreaking as it was, did not fit the model of academic tenure in my department. I left academia and went into consulting. I would eventually come back to academia as an adjunct professor. And, between 2007 and 2013, I ran the online advocacy program at the Center for American Progress, where I developed CAP’s social media strategy (with my colleague Brian Komar) and built/managed the program to implement it. 

So, one of the big changes I have seen is that the Political Science community has finally embraced the digital political revolution as real. I say that in jest, because while they came to that position years ago, it was nowhere near as soon as they should have given the mounting evidence.

In the early days of digital politics, it was all about websites, then it was about ads on websites. Websites used to be like online brochures. But as web apps evolved, the functionality of websites improved. By 1996 we could develop campaign websites that matched people to their elected officials, allowed them to send those officials emails, and track all the data associated with that mobilization with an off-the-shelf tool called Capwiz.

While many called the internet an information revolution, I always considered it a communication revolution, and information was just the content we shared. That mindset made it easy for me to think about social media strategy long before anyone was calling anything online “social media.”

So as we moved from USENET, IRC, and CU-SeeMe to Friendster, MySpace, and MeetUp, and ultimately to Facebook, Twitter, Instagram, SnapChat, YikYak,  and TikTok (should I go on?), I saw that with each new set of tools, the basic social media strategy remained the same. Get your message out, refute the other side’s message, and recruit supporters. But the tactics evolved with each new set of tools.

I also saw that in the early days, the most effective users of these strategies were white supremacists and terrorists. Over the years, the good guys got much more effective, catching up to the bad guys, but it remains an arms-race, as the current battle of the bots rages on.

Another change I advocated for over the years and am increasingly seeing in the advocacy space is the use of social media to deliver messages to lawmakers. Generally, email joins phone calls, letters, and personal visits as the dominant channels for communicating with Congress. But in the past few years a handful of advocacy platforms are also delivering messages via Facebook and Twitter. And, too slowly for my taste, we are seeing tools being deployed on the receiving end in Congress to monitor what people are saying to lawmakers via social media. Now all we have to do is make sure lawmakers can tell the difference between citizens, bots, and foreign agents. That tech exists but is insufficiently deployed. 

2. Does any of your research suggest that social media has increased the number of informed American voters? 

I have seen research from Pew that shows people are paying more attention to elections many months earlier than before the internet. Instead of waiting until two weeks out, they are tuning in five to six months out, or more. Pew also found that at least a fifth of voters say social media influenced their position on a candidate or an issue, and voters say they learn more about candidates via their social media than via their email and website combined.  

Generally, social media informs a vast number of voters. The problem is that many are falsely informed. We always think that information is necessarily true, but that is not the case. The mythical “informed electorate” may be more informed about what the opposing candidates say is true than about what is actually true. Politics has always been about persuasion and perception, but with social media we are seeing that persuasion and perception is not a guarantee of justified true belief (an old epistemological standard for evidenced knowledge).

But while social media has given rise to the super-informed, but not necessarily factually informed, electorate, ensuring that we all work from the same belief in facts remains a challenge; I suspect it always was, but social media shines a bright light on it. That said, I believe that social media and related tools exist to empower everyone to be factually informed… potentially.

3. How can we combat the use of media to spread news that is unreliable in hopes of political gain or that sensationalizes events in attempts to stir a public reaction?

This is not an easy task. There is no one solution, but rather a suite of solutions that all must come together. Most people begin with improving people’s media and information literacy. This is important. And it fits the mindset that we are all responsible for our own lives. But it cannot solve the problem on its own. Technology can fool the human senses. 

On the tech side, ensuring that there are effective ways for people to report content they suspect is fake is crucial. The current reporting categories on Facebook and Twitter, for example, are inadequate to this task. Using AI to help flag suspicious content is also important. But in both cases, it ultimately requires humans to verify the report. And that can be a tedious and depressing job, as the horror stories from Facebook’s review teams indicate.

There are also several projects being funded by The Craig Newmark Foundation, Facebook, Google, the Knight Foundation, and others around the world to build out the capacity for journalists to verify information spreading around countries and the world. This is crucial work, but in the end, it will only be successful if they do it well and if we are able to restore the people’s trust in journalism.

On the regulatory side, it is a constitutional nightmare in the US. The right to free speech, which is among the first amendment rights that the second amendment was written to protect, makes it difficult, at best, to censor political content. Twitter’s policy of not enforcing its Terms of Use regarding disinformation and abuse on world leaders is a case in point. If regulations created to restrict/police disinformation were applied fairly, we would see our President banned from social media as a result, and that might cause serious strife across the country. But there seems to be a need for careful regulation, at least with respect to accountability and consequences.

4. Has your research suggested ways to most effectively communicate a message through media? How can a campaign use their data and analytics to improve their social media strategy and increase the impact of their media messaging? 

Effective social media strategy to communicate your message requires that you 1) have a compelling message, 2) test it on agenda-setting channels like Twitter and Reddit to see if it gets traction, and 3) push the best message out over multiple channels via multiple voices. The days of responding to an issue or a scandal with a press release, a press conference, and few media interviews are over. Those methods have a shelf life that expires as the social media amplification chambers starts to explode.

Many people look at social media and think it is an echo-chamber, and at times it is. But effective messaging ultimately breaks out of the echo chamber and is amplified in many directions at once (making it an amplification chamber). It is like a bullet fired inside a room surrounded by nearly bullet proof glass. It ricochets until each wall gets just enough kinetic energy to explode outwards.

5. How impactful is the use of social media during political campaigns if the target audience skews older? 

I don’t think people assess the impact of good social media strategy correctly. Across all ages there are people using social media. And everyone who uses social media influences people offline, as well as online. When a parent learns something on social media, they may discuss it with their family at dinner. In the workplace, older and younger people interact, talking about what they have learned from whatever sources they consume. Social media online is part of the social graph and the social graph is everywhere and fluid. Information enters the system at many points, flows to many people on the initial network and then jumps to other networks, online and offline. 

A few years ago, Buzzfeed developed a tool that allowed them to track shares of their stories on social media, both downstream and as it jumps from one network to another (like Twitter to Facebook). Even without including the jump to offline networks, they saw significant cross-platform sharing.

If we use the internet as a metaphor here, where information is broken up into packets and distributed in every direction at once to be reassembled on the other end, political campaigns should be using all of their channels to push out their messages, to refute the other side’s message and to recruit supporters. Voters will reassemble the messages they get from all these channels on the receiving end. When campaigns use multiple channels and multiple messengers per channel, the impact is greater than the sum of the impact from the parts.

6. What is your best tip for campaigns that are struggling to reach their target audience through online mediums?

Influence the influencers. Before you can influence the masses, you need to influence the influencers to spread your message to their audiences. As you build up this capacity, your message not only reaches more people, but it reaches them through multiple channels and from multiple people that they already trust.

7. What is the worst/most common mistake you see politicians or advocates making when trying to employ social media to engage voters?

I hate it when politicians say “read this article, it is important,” but don’t tell you what it says and why it is important. Most people will never click on the link, so if they read the tweet and did not click, the politician has left a voter on the table without delivering them their message.

Similarly, I hate when politicians tweet a photo of themselves with a supporter at an event and say how much they enjoyed talking with so and so, but do not tell us what they said to so and so. Again, this is an opportunity to market your ideas, to say, 

“I met Sandy at the Carrier factory in Indiana and we talked about how all those jobs Trump promised to save at the factory went away. Then we discussed how clean energy investment can replace those jobs.” 

That is so much more powerful than saying, “I had a great conversation with Sandy at the Carrier factory in Indiana. Thanks for taking the time to talk with me.”

Finally, I think politicians misunderstand social media strategy and think it is just about broadcasting, when it is also about organizing. Time to change that misconception.

BONUS QUESTION: Can social media mobilize potential young voters to the polls? 

Yes, but not if you try to guess which platform will be best to reach them. Use ads to reach all channels, apps, and websites, regardless of platforms, and then engage people where they reply. Then follow up with peer-to-peer GOTV mobilization.

SECOND BONUS QUESTION: Have you read any books or listened to any podcasts lately that you would recommend to our readers? 

Joe Romm’s How to Go Viral and Influence Millions, available only on Amazon, is fantastic.

Thank you so much to Dr. Alan Rosenblatt for sharing with us his social media strategy wisdom!

Questions? Email Us
Categories
Receive EMAIL UPDATES from The Campaign Workshop blog!

Lastest Blog Posts

The Resist List
Resist List 2019: fist with colorful confetti flying off of it