Another Tool in the Influencer’s Toolbox: A Case Study
By: LTC Jamie Efaw and SFC Chris Heidger
Much of the recent writing within military circles surrounding social media has centered on what can be learned from "bad guys'" use of social media sites, or the role social media played (and play) in events such as the Arab Spring and Haiti disaster relief. However, from an information operations perspective, the "golden ticket" in our operations is the ability to influence behavior and attitudes. The questions we address in this article are: "Can an information operator use social media to influence audiences, and if so, how?" Attitudes and behaviors cannot be changed overnight; doing so requires exposure to a persuasive message repeatedly over an extended period of time—a task social networking tools are perfectly designed to accomplish. One should not, however, view social media as the single best way to exert influence, but rather, see online networking as another tool in the communicator's toolbox that enhances larger influence campaigns. In this paper, we will share some of our experiences from putting theory into practice, our phased approach, some lessons learned, and recommendations for getting started in your own organization.
Getting Started and Building the Target Audience
In the winter of 2008, I (Jamie Efaw) published a paper in IO Sphere titled "Social Networking Services: The New Influence Frontier."1 I had been thinking about these ideas for quite a while, and at the time, the paper was largely theoretical, focused on the "whys" and "hows" of social networking platforms and their application toward counterterrorism efforts. In 2009, my new NCOIC (non-commissioned officer in charge) and co-author Chris Heidger and I arrived at the United States European Command (EUCOM) in Stuttgart, Germany at the same time. We found we had inherited an eightyear old regional Web news and information initiative called the Southeast European Times or SETimes, which had an accompanying eight-month old Facebook companion presence.2 We both saw this situation as a perfect opportunity to leverage social media and apply the theoretical framework I'd developed in my paper to a real-world situation. As a starting point, we identified four goals for having a social media presence:
- Take advantage of the existing social media community to introduce and draw them to SETimes.com.
- Provide an additional forum that exposes our target audience to our themes and messages.
- Provide a convenient place for the SETimes community to discuss regional topics of interest and interact in an environment where they are comfortable and familiar.
- Establish a communication platform we could use during a crisis, humanitarian assistance, or disaster relief operation.
In an effort to minimize the risk of our enterprise failing to catch hold or committing a public relations foul, while simultaneously maximizing learning and flexibility, our social media expansion efforts took a very deliberate five-phase approach, which is laid out in detail below:
Phase 0: Establish a Facebook Presence
Phase I: Research and Improvement
Phase II: Focused Advertising
Phase III: Increased User Interaction
Phase IV: Shift Focus onto Established Social Networks
Figure 1 summarizes the SETimes' Facebook fan growth by phase from March 2010–January 2011.
Phase 0: Establish a Facebook Presence (March 2009– February 2010)
Phase 0 started prior to our arrival at EUCOM, and consisted of creating a Facebook page for SETimes, adding a RSS feed from the webpage to the Facebook page, and putting a Facebook link at the top of the SETimes homepage.
During our first six months on the job we made no changes; noting that readership ("fans") and interaction (fans "liking" a post or making a comment) on the Facebook page did not increase over that time, however, we realized that in preparation for Phase 1 we needed to benchmark the current initiative. We do not see ourselves as competitors with the Department of State (DoS), but we used official DoS Facebook pages to serve as a yardstick for our site, since they have a comparable presence around the globe. A year after its creation, the SETimes' Facebook page gained an average of less than one fan per day, for a total of 306, and ranked in the bottom 15% when compared to similar DoS pages. Our audience was largely U.S.-based, and there were more English-speaking fans than all other languages combined (see Figure 2).
Phase 1: Research and Improvement (March–June 2010)
In March of 2010, we began our research and improvement phase, which started with the question: "What is the dominant social media platform in Southeast Europe?" A good surface-level tool for this research is Alexa.com,which identifies the most popular sites in a targeted region. With Facebook rated as the frontrunner (which will not always be the case), we then evaluated successful Facebook presences similar to SETimes, such as U.S. government pages and those of other regional news organizations, to get an idea of what they were doing right.
Based on our findings, we took a few minor steps to improve our Facebook page. First, we opted to manually post content in lieu of RSS feeds, which for a social platform we decided were too impersonal, mechanical, and annoying. Second, we added new features such as photo albums and a discussion tab with conversation topics, in an effort to promote a social atmosphere that was more in line with Facebook's interactive environment. Third, we published a story on SETimes.com about Facebook in Southeast Europe, which also encouraged readers to join our Facebook page (see Figure 3).3
Lastly, in May 2010, we provided basic Facebook instruction at our annual SETimes writers' conference, and encouraged our contributors to become "fans" and interact on our page.
Despite focusing on Southeast European audiences, Figure 4 shows that we still had a very large U.S.-based audience and an even larger ratio of English-speaking fans than we had at the end of phase 0. While we understood that we still had a long way to go, we were encouraged that our new fans per day had doubled, indicating an increase in the page's appeal.
Phase II: Focused Advertising (July–October 2010)
Phase II, which started in July 2010, is where our project started gaining significant momentum. The first noteworthy change was the introduction of advertising to Facebook users in SETimes' twelve core countries (see Figure 5).
We chose a rate of $5/day, which made Facebook advertising a cost-effective as well as user-friendly tool. Getting started requires only a credit card, some small thumbnail images, an introductory sentence, and an intended audience. Once established, advertising costs are set through a daily geographically- dependent bidding process that ranges from less than U.S. 10 cents to more than $1.00 per ad click.
Demographic data are the greatest advantage that Facebook provides a planner. These data include age, sex, country, city, marital status, education level, and even interests, all of which are gathered as users establish and update their profiles. These are the data that focus advertising and provide excellent, real-time, demographically separated measures of performance. Although we have used Facebook ads only to increase the number of fans on our Facebook page, these ads can satisfy other objectives such as advertising an event, directing people to a website, or simply getting a target audience to download an application.
While advertising was an important step, we also added some additional features to enhance the users' overall experience during this phase. First, we added the "Like" button to individual articles on SETimes (Figure 6).
When SETimes readers who are also Facebook users click that they "like" something, the action is indicated on their Facebook pages, and shows up on their friends' pages, along with a link to the "liked" content. Second, we added prompts to our Facebook posts in order to encourage interaction. Typically, we would link to a recently published SETimes article and then post a related question or comment. For example, we posted an article about the start of the Croatian president's second year in office, along with the following question: "Josipovic started his 2nd year as President of Croatia. How do you think he did in his 1st year? Share your thoughts."
Throughout Phase II we saw a sharp increase in the growth of fans from our target region (see Figure 7). Due to focused advertising, we now gained a daily average of 35 new fans, 99% of whom were from our target audience. As also depicted in Figure 7, U.S. readers dropped off the readership chart, and English language usage as a proportion diminished.
After four months of posting content to our page once every five days, we realized that Facebook users tended to visit pages other than their own only on the day new content appeared on their wall. The "if you build it, they will come" model, in other words, does not apply in social media. In the world of "new media," consumers no longer actively seek out or pull information; instead, they subscribe to topics of interests and have the information fed to them. In Phase III we adjusted our approach to fit this model.
Phase III: Increased User Interaction (November–December 2010)
Phase III began in November 2010. The goals in this phase were to implement lessons learned from Phases I and II, increase fan interaction, and test some of the more specific advertising techniques.
Observations from previous phases revealed that the timing of Facebook posts affected interaction levels. For two weeks we logged hourly Facebook traffic from 0600–2300. This showed us that peak traffic for our target audience occurred between 0900–1000 and 1700–1830. Based on these findings, we began posting to Facebook twice a day, during each of these periods. We reasoned that this technique would optimize views and reader interaction while minimizing the risk of becoming a nuisance. To further encourage interaction, we started producing exclusive content for our Facebook page such as video essays, and we introduced a weekly "hook," such as a survey question, a "complete the sentence" statement, a "fan of the month" post, or a quote of the week.
Other early engagement efforts taught us several useful lessons. First, heavier content, such as the re-posting of a SETimes feature article, performed better in the morning, while lighter posts received better traction in the evening. These lighter interactive evening posts (often devoid of core themes and messages) were essential to keep the page in line with the social nature of Facebook. Finally, we found that simple encouragement such as, "Thanks for your comment, we value your insight," or "Interesting photo, can you tell us more about it?" proved invaluable as a method to keep fans engaged and let them know we were paying attention to them.
From our advertising campaign, we observed another interesting trend. During Phases I and II, audiences from our lower-priority countries (based on existing national and command guidance) quickly used the majority of our daily advertising budget by clicking on our ads more frequently than higher priority-country audiences. To rectify this, we divided our increased advertising budget of $10/day in accordance with SETimes' established threetier priority system. Tier 1 (highest priority) audiences received $5/day, while Tier 2 audiences received $3/day, and Tier 3 received only $2/day. Lastly, due to the steadily increasing number of fans and interaction, we anticipated a need for a Facebook Terms of Service statement that would provide general information about SETimes, and codify acceptable and unacceptable member conduct.4 Our Terms of Service proved useful later when dealing with inappropriate reader comments.
Until this phase, EUCOM command policy was to block all social media sites, so all these Facebook activities had to be managed from non-government computers. After months of dialogue with communicators across the staff, Admiral James Stavridis, the EUCOM commander, published a comprehensive social media policy granting command access to social media sites. In addition to making our Facebook campaign far more convenient to maintain, workplace access enabled us to determine peak traffic times, increase posting frequency, and continuously monitor our Facebook page— maximizing the efficiency of a real-time communication platform. During Phase III, English-language users continued to drop off, while fans who used SETimes target-country languages continued to increase at a rapid pace (see Figure 8). While our advertising doubled, there was, surprisingly, an increase of only four more fans per day compared to our daily average in Phase II—a trend that has not changed. In fact, the return on investment dropped off very rapidly as daily advertising levels increased. For example, if a Facebook page gets twenty new fans per day for $10, it does not necessarily mean that investing $50 per day will bring in one hundred fans. Phase IV: Shift Focus to Established Social Networks (January 2011–ongoing)
At the beginning of 2011, we initiated Phase IV, which involved increasing the advertising budget to $100/day and focusing our efforts on increasing four measurable Facebook page user actions: 1) page visits; 2) page "likes"; 3) fan interaction; and 4) referrals to SETimes.com (i.e., reads of SETimes.com content). In order to promote an increase in the above actions, we recognized the need to leverage the power of social comparison that is inherent in the Facebook platform. Social Comparison Theory postulates that people evaluate their opinions and actions based on their peer group. When one's own opinions or beliefs are in contradiction to the comparison group, there is a tendency to either change one's own opinions or attempt to convince others that your beliefs are correct in order to re-establish group cohesion. Typically, when encountering content on a traditional website, the reader does not know what other friends have read the same article, who liked it, or what they thought about the content they read. Social media platforms, however, are by definition built on social connections, and thus are formatted to allow the reader or fan to engage in social comparison.5
In previous phases, we advertised to all Facebook users within SETimes' established target audience (over 30 million Facebook users, and growing). This all-encompassing approach established nodes in existing social networks in every corner of our focus countries; however, the majority of our fans were not connected in any meaningful way. To rectify this, we expanded established nodes (existing SETimes Facebook fans) and began advertising only to their Facebook "friends." Currently, a tag at the bottom of every SETimes Facebook ad might say something like, "Joe Smith likes this." As more "friends" within a circle "liked" the ad, it might say "Joe Smith and 5 of your friends like this"—increasing the power of social comparison with every new fan.
These newly formed pockets of SETimes fans not only increased the effectiveness of our advertising, but also increased the potential of both online and offline conversations with friends surrounding the content of the Facebook posts. Figure 9 illustrates that previously observed trends continued, with a marked increase in the growth of fans from the target audience.
Once a social media user joins the site or page, social comparison continues to have an influence. When readers interact with content on Facebook, they immediately know how many other total users have "liked" the content, but more importantly, they know how many and which of their own friends (their comparison group) "liked" it (they do not see which of their friends do not like an article). Additionally, they can see any comments made about the content and make comments of their own in real-time. Whether the reader tries to change the opinion of friends or changes his or her own to match theirs, either outcome could be useful for the influencer if properly managed.
Lessons Learned and Difficulties
Over the last eighteen months we have identified several advantages, difficulties, and lessons learned while using Facebook, many of which already have been discussed above. The following section offers nine broad topics we consider principles for getting started.
1. Research and listen.
Where are the conversations among your target population taking place? On what platforms? What are the advantages and disadvantages of the different social media platforms for reaching your audience? Answer these basic questions before launching a social media effort.
2. Have a plan, but experiment. Be deliberate, expect failures, and be flexible.
Before starting, have a plan and be deliberate in your method, but also leave room to experiment. The same principles that worked for us will not necessarily work for everyone. As you experiment, start small, expect failures, and be prepared to make adjustments. This will ensure you have the flexibility to make required changes while minimizing the negative effects of failures and growing pains.
3. Provide fans with experiences. Make interaction easy. But do not saturate!
We found that to get fans involved with our content, we needed to provide online "experiences" that invited them to interact. An activity that encourages engagement, participation, and conversation, yet remains easy and quick for the reader, is ideal. In the social media world, more is not always better. A user typically is not going to join in an in-depth erudite dialogue, but most will likely vote in a poll, complete a sentence, or "like" a quote. Too much information (i.e., too many posts) hinders your effort. At best you run the risk of being ignored, and at worst, annoying the individuals you want to influence.
4. Benchmark your efforts and progress.
Benchmarks help you see progress toward your goal. If you do not establish a reliable baseline from the outset of your activity, it is impossible to show how far you've come at subsequent benchmarks, or at the end of your effort. In March of 2010, SETimes' Facebook page had 306 fans and was ranked in the bottom 15% when compared to similar DoS Facebook pages. Less than a year later, we had over 17,300 fans and it was ranked in the top 6% compared to similar pages. Other aspects can be benchmarked as well. Our Phase IV measurable actions provide a good starting point: 1) page visits; 2) page "likes;" 3) fan interaction; and 4) referrals to an external website (i.e., reads).
5. Add a Terms of Service policy.
A Terms of Service policy outlines acceptable and unacceptable user behavior on the site. By stating policies upfront, you have de facto authority to delete inappropriate posts and ban disruptive users from the site. On several occasions, this policy enabled our fans to regulate each other and make sure that fellow fans abided by the spirit and the letter of the policy. This was also a welcome indication to us that our page's fans were taking "ownership" of the page and regarded it as theirs to monitor.
6. Educate the organization.
Much can be written about educating your organization. Be aware that some individuals will never understand, or desire to understand, what you're doing and how you're doing it. However, there are others who are willing to learn. Help these people set up an account and show them how social media work, then encourage them to experiment from their personal computers. Ensure you are an available resource to answer questions, but be careful to do it without making anyone feel stupid. Locate and mobilize key personnel in your organization. If possible, find senior leaders in your organization who "get it" and who will act as advocates when needed. Similar to the "start small and expect failures" discussion above, do not make unreasonable or unrealistic promises to your bosses. We kept our developmental phases pretty quiet until we were able to demonstrate some solid metrics. Furthermore, because we started small, there was no need or requirement to report setbacks.
7. Real-time interactions.
Real-time interactions are both Facebook's greatest advantage and its greatest challenge. The challenge lies in the fact that these interactions require constant monitoring in order to respond appropriately. Social media encourage conversation, which assumes timely responses. Failure in this area alienates fans and discourages future participation. A unique advantage of real-time interaction on Facebook, however, is that every time a fan comments on a discussion thread, the comment is sent out to everyone who has previously taken part in that thread, as well as all of the friends of the commenters. This feature continues to draw people back into the dialogue.
8. Translation and the use of an official language.
It is important to be aware that social media tools are ideally suited for campaigns focused on a single audience with a common language and problem set. Our audience, in contrast, spans twelve countries and ten spoken languages. Regardless of language(s) used, there are two translation management issues: 1) fan comments submitted in a language other than the page's official language(s); and 2) the general difficulty of maintaining oversight on a second-language (for the manager) fan page. For the first issue, on-call translators are required; this, however, can take time. While waiting for full translation, we have found the Google Translate and Live Translate applications to be useful for an immediate base-line understanding. To maintain oversight of a second-language Facebook page requires a minimum of one full-time, native speaking-level employee.
9. Leadership's lack of understanding.
There are many people who view Facebook as an invasive site that reveals far too much information about its users. Although mistrust was a more common problem three to four years ago, it still is an identifiable generational issue and a recognizable hindrance to social media initiatives within an organization.6 While many senior-level leaders still do not fully understand social media, most at least understand its importance. Basic understanding and acceptance is all one needs to move forward.
Summary and Conclusion
When we look back at the original four goals we set for getting involved in Facebook, we are gratified to see that we achieved each of them.
Leverage the existing social media community to introduce and draw them to SETimes.com.
By the middle of 2012, we will have accumulated nearly 400,000 SETimes Facebook fans. Nearly all of these members of our target audience are individuals who had never heard of SETimes.com, nor been exposed to the website's themes and messages prior to the launch of our Facebook page. Additionally, in 2011, we recorded over 120,000 visits to SETimes.com from readers who came directly from our Facebook page; in other words, they were on the SETimes' Facebook page, clicked on a hyperlinked headline that interested them, and then read the full article on SETimes.com.
Provide an additional forum that exposes our target audience to our themes and messages.
If a fan never clicked on one of the Facebook-posted links taking them to the SETimes.com content, they would still be exposed, three times a day, seven days a week, to its themes and messages via postings and interactive features on the SETimes' Facebook wall.
Provide a convenient place for the SETimes community to discuss regional topics of interest, and interact in an environment where they are comfortable and familiar.
We assessed our achievement of this goal by asking our fans, in a Facebook poll on January 23, 2012, what they liked most about the SETimes' Facebook page. The great majority (72%; 1720 of 2390) of respondents acknowledged that the feature they appreciated most was that the page provided "a forum for an exchange of opinions and dialogues with others in the region" (see Figure 10).
Establish a trusted communication platform that can be leveraged during a crisis, humanitarian assistance, or disaster relief operation.
This function was validated on several different occasions. A good example occurred in September 2011 during a flare-up of tensions at the Serbia-Kosovo border. Utilizing the SETimes' Facebook page, we were able, first, to inform the target audience of the actual situation on the ground, thereby discrediting rumors and disinformation. As a result of this effort, combined with our Google advertising, SETimes' coverage of the event was the #1 overall search result in Google news. Second, by utilizing the Facebook polling option, we were able to quickly gauge sentiment among the target audience, and provide that information to commanders and decisionmakers. The fact that our audience trusted the integrity of our information was key to our success on these occasions.
The question posed in the first paragraph of this article, you may recall, was: "Can an information operator use social media to exert influence and if so, how?" After two and a half years of daily experience with the SETimes platform, it is evident to us that Facebook and other social media sites are valid influence tools. First, social media allow an influencer to expose the target audience multiple times to the crafted content. As one can observe in Figure 11, over one million Facebook users were exposed to (had read or viewed) SETimes Facebook content more than five times in a seven-day period.
One reason this statistic is encouraging is that users not only saw SETimes content several times, but they chose to have our content delivered to them or they actively sought it out.
Second, we discovered that appropriately designed content prompts the target audience to share the message with others. Participation in a dialogue or a conversation about a topic encourages an individual to think deeply about the topic. To see how often this was occurring, in December 2011, we asked the poll question: "How often have you had a discussion with someone about a topic/issue you read about on SETimes?" (see Figure 12). Eighty-nine percent (2099/2365) of respondents acknowledged that they had had conversations about SETimes' content either daily or weekly.
These results confirmed that our target audience is thoughtfully considering the material presented by SETimes. The results also indicate that the online target audience is expanding the reach of the SETimes' message by having conversations with their off-line, real-world social network, thus expanding SETimes' sphere of influence.
Lastly, while we have illustrated that well-designed social media offer the tools to deliver a message and encourage dialogue, measuring influence (effectiveness) still remains difficult. Besides observing a change in the behavior of members, self-reporting (polling) is often the only method for determining effectiveness. To this end, we asked our Facebook fans: "Has anything you have read on SETimes.com caused you to think differently about an issue?" (see Figure 13). Although people are often reluctant to admit they have been influenced, an impressive 93 percent (2693/2916) of SETimes' Facebook fans who responded acknowledged they had been affected by our product. As one can gather from our experiences, social media initiatives require constant attention; the good news is that they require only limited personnel and funding to have an impact. No matter how enticing this high return on investment may sound, however, it is essential to keep in mind that Facebook, or any social media platform, is not a stand-alone tool. In fact, it will always work best in concert with traditional communication tools supporting a larger effort, event, or cause, such as, in our case, the SETimes news and information website. Used properly, this rapidly evolving capability is proving itself a tremendous addition to the communication toolbox.
About the Author(s): LTC Jamie Efaw serves as the Fort Story ASA Commander and Deputy Commander for Joint Expeditionary Base Little Creek-Fort Story in Virginia Beach. He was commissioned into the Corp of Engineers by the United States Military Academy at West Point in 1993 with a degree in Psychology. After receiving his Master's Degree in Social Psychology, LTC Efaw went on to teach in the Behavioral Science and Leadership Department at West Point, and branch transferred to Psychological Operations. After serving in the 4th Psychological Operations Group at Fort Bragg, North Carolina, LTC Efaw served as the United States European Command's Military Information Support Operation officer. From 2009–2012, SFC Chris Heidger was the Senior Enlisted Military Information Support Advisor at the U.S. European Command in Stuttgart, Germany, where he and LTC Efaw managed Southeast European Times. A nine-year PSYOPS veteran, Chris has spent five years overseas with tours in Iraq, Afghanistan, Africa, and, most recently, Europe. Chris earned a BA in Political Science from American Military University, and recently finished the coursework for a Master's of Professional Studies in Strategic Public Relations at George Washington University.
1. James M. Efaw, "Social Networking Services: The New Influence Frontier," IO Sphere (Winter 2008): 4-7; http://www.au.af.mil/info-ops/iosphere/09winter/iosphere_win09_efaw.pdf; accessed on September 27, 2012.
2. The Southeast European Times website is a central source of news and information about Southeastern Europe, offered in ten languages: Albanian, Bosnian, Bulgarian, Croatian, English, Greek, Macedonian, Romanian, Serbian, and Turkish.
3. Natasa Radic, "Facebook has a friend in Southeast Europe," SETimes.com, March 8, 2010; http://www.setimes.com/cocoon/setimes/xhtml/en_GB/features/setimes/articles/2010/03/08/reportage-01; accessed on October 1, 2012.
5. Leon Festinger, "A theory of social comparison processes," Human Relations vol. 7, no. 2 (1954): 117-140.
6. Meagan Johnson and Larry Johnson, Generations, Inc.: From Boomers to Linksters—Managing the Friction Between Generations at Work (New York: AMACOM, 2010).