I met Andrew Gossen at the 2011 CASE Social Media Conference in San Francisco. He was a presenter at the conference, and his session walked us all through the importance of social media monitoring. At that time, I thought Andrew was a man ahead of his time, and he still is. He leads the way in social media practice at the higher ed level nationally, as well as at Cornell. In this interview, I asked him to talk a little about his personal experience with social media monitoring and the role it plays in higher ed communications strategy. His common sense approach to social media is refreshing and innovative. A little known and interesting fact: Andrew has Ph.D. in social anthropology. Read on to find out how that influenced his decision to get into alumni affairs, and ultimately, social media. My sincere thanks to Andrew for taking time out of his busy schedule to spend time with us.
1. How did you get involved in social media?
When I joined Princeton in 2002, there was an online education component to the alumni education program. We were investing time and energy in videotaping academic lectures and making them available to alumni for viewing online, as were many institutions at that time. It just wasn't working, and I was growing increasingly skeptical that online engagement with alumni would ever be viable. In 2007, however, I stumbled across a very early example of social gaming, a start-up called GoCrossCampus that was running an online Risk tournament among the Ivies. This competition involved both students and alumni, and the amount of activity that I saw forced me to re-evaluate everything. Alumni clearly wanted to interact with each other online; we just weren't offering them anything that they wanted to do. This realization, combined with an explosion of alumni activity on Facebook, made it clear that the ground had shifted in a very important way: alumni now had the tools to organize themselves around the things that they cared about. Not only was there no need to get permission from alumni affairs; they didn't even need to let staff know what they were doing! Up until this point, no large-scale alumni activity could happen without alumni affairs because they controlled the database. That monopoly had been broken, and I concluded that if we didn't recognize this and adapt, we were going to be left behind.
I decided to dive in to this emerging world. Much to my surprise, I found that working on the social web actually dealt with many of the issues that originally had attracted me to anthropology. What causes groups to form? What catalysts cause communities to mobilize? What's the relationship between individuals and groups, and how do groups relate to each other? On the social web, you can watch these dynamics unfold in real time, happening millions upon millions of times as the user bases grow on the various social platforms. Having the chance to work on these issues in an applied way, figuring out howto cultivate and mobilize online communities on behalf of Cornell, has been challenging, rewarding, and a whole lot of fun. Cornell alumni are an amazing group of people, and it's a privilege to work with them in this way.
2. At the 2011 CASE Social Media Conference in San Francisco, you presented a case study of an online event that impacted Cornell, and how you used a social media management system (SMSS) to track the event as it unfolded on the social web and make a decision as to whether or not you would respond. In your thinking, how important is it for institutions to be monitoring (listening to) conversations about their brand online?
It's of paramount importance. One of the questions I hear most often from people who are deliberating about whether or not to jump on the social bandwagon is this: what if people say bad things about us?That's the wrong question. People are already saying bad things about you. The real question is whether or not you want to know about it. Sure, ignorance might be temporary bliss, but you could be setting yourself up for nasty surprises if you don't deal with the reality that conversations about you are taking place on the social web, whether or not you like it.
It would be wildly irresponsible to ignore this. You want a smoke detector in your home to alert you as early as possible when you have a problem, don't you? Monitoring your brand plays a similar role. If people are angry about something and it's spreading, don't you want to know? If someone is impugning your institution's reputation by spreading rumors or facts that are incorrect, don't you want the chance to rebut them? Mind you, knowing about something doesn't obligate you to do anything about it. You can assess the situation and make a decision about whether and how to get involved. But you can't have any sort of a stance on something you don't know about. By not monitoring your brand, you may be confronted with a crisis that's much more intense than it would have been had you acted earlier, and people will be surprised. And in my experience, people in higher ed don't deal gracefully with surprises.
3. Obviously, you had the monitoring software in place before the online event. What process did you go through in designing an online listening program?
We evaluated a number of vendors and went with the one that provided the right combination of ease of use, monitoring breadth and depth, the ability to set up campaigns monitoring custom combinations of phrases and keyword, and good analytics. It was total serendipity that we acquired the tool about six weeks before this event took place - it would have been much more difficult to figure out what was going on otherwise.
4. In your presentation, you shared that you ultimately chose not to enter the conversation about this particular event. Can you describe the response strategy you used for this event? Did you have a social media crisis response plan, or did you learn as you went?
The overall response to the incident involved a team of communicators from across campus. In some high-visibility contexts, mostly print, we did respond to provide context, a more complete and nuanced description of the event in question, and corrections of blatant errors and misrepresentations. My role was to monitor the story as it moved into the social web, mostly Twitter and the blogosphere, and provide a response when and if it seemed appropriate.
Over the course of four days of close monitoring, the initial spark of activity gradually abated. In the few instances when the story made it to online destinations with sizeable audiences - The Huffington Post, for instance - the audience didn't seem to engage. People weren't leaving comments and they weren't sharing it onward, clicking "like" buttons, or doing anything else that indicated that it was picking up traction. Since it seemed like the problem was dying out on its own, it made more sense to let it do so than risk sparking rekindled interest by scattering official response to every mention of it throughout the social web.
5. What did you learn from this experience?
I was interested in how people responded to this case study at the Social Media and Community conference. It's so rare for anything to actually go viral on the social web that it seemed to me like agood gamble to let the story fade out on its own - the lifespan of content can often be measured in seconds. However, I did talk to a number of people in the audience, most of whom had backgrounds in print journalism, who felt very strongly that I should have provided a response every time I found the story mentioned on the grounds that no misrepresentation of the facts should ever go uncontested. Having watched similar crisis management processes unfold at other institutions, it always struck me as too much when a university representative would provide a three paragraph response to a throwaway post on an obscure blog with 15 subscribers. A completely disproportionate response can actually work against your best interests by making people suspect that there must be something to the story.
Don't get me wrong - if you see a problem taking root in a place where there's a high likelihood that it's going to spread, or if you can see it spreading, you'd better be on top of it rapidly. But it's really important to understand the social web ecosystem well so that you can differentiate between a burgeoning crisis and a situation where engaging might actually make things worse.
6. What is your advice for other organizations struggling with the labor and cost of building an online listening program?
Cost shouldn't be an issue. While you can spend hundreds of thousands of dollars on a custom monitoring package designed with Fortune 500 clients in mind, you can also do just fine with a combination of HootSuite or Sprout Social and Google Alerts.
The labor is a different - and underappreciated - factor. You can have the most elegant suite of tools available, but if you don't have time to look at the data, you might as well have done nothing. Some tools can automate part of the monitoring process, sending you summaries by e-mail of the activity on keywords that you're monitoring. Others have features that alert you if there's a sudden spike in activity that the system thinks is negative. But even these shortcuts don't do you any good if no one reads the e-mails.
At the bare minimum, you should have a monitoring tool in mind to use should you suddenly find yourself in the middle of a social media crisis. On a slightly higher level, you ought to be monitoring the name of your institution on Twitter, at least - intense activity in other areas of the social web tendsto bleed over into Twitter. You don't have to watch it all the time, but taking a look once daily would be nice. And at the opposite end of the spectrum, think long and hard before making a significant investment in one of the higher end tools. Does your institution generate enough traffic on the social web to need this option? I once did a comparison of mentions of Cornell to mentions of the iPad2. During a two-week period, the tool picked up an average of 600 mentions of Cornell each day. Mentions of the iPad2? An average of 180,000. For Apple, this tool was a necessity. For Cornell, it would have been a luxury. Be realistic about your needs, and the resources you have to process and respond to the data you capture, and you'll arrive at a solution that's right for you.