Lessons Learned: Joplin Crisis Communication

This has been a bit long in coming, I know. One month ago, on May 22nd, an EF5 tornado tore one of the most destructive paths in modern history through the middle of Joplin, MO. As of today, 155 fatalities have been attributed to it. This is not new to anyone at this point. Those of you familiar with me on Twitter or Facebook no doubt took note of the small role that I played in helping with communication during that chaotic time. I learned a lot from that. A whole lot. It’s taken some time to go through it all in my head and sort things out, and this past month my family faced its own tragedy in the passing of my grandfather. Needless to say, this was something that needed to get written, but I needed to get my head in the right place first.

First off, with great luck our friends at MSSU managed to come through pretty much unharmed. They mobilized relatively quickly to use their available space for shelter and triage. Kudos to all the fine, hard-working folks there. I know many were worried early on, and it didn’t come without some cost. We must always remember, universities are large families with our faculty, staff, students, and alumni. In a large scale event, we will be impacted somehow, it is simply unavoidable. Whether tornado, hurricane, earthquake, etc, nearly all of us should at least consider the possibility of a major event and what we will do both as organizations and individually.

On the night of May 22nd, I was sitting quietly in my recliner, no doubt doing something unseemly on Facebook. My wife and I had the TV on as the news broke in right as it hit (at the time, they didn’t realize what it was they were catching besides some windswept debris). Within minutes, I started hearing some chatter on Twitter, and before long, the news started getting some word about the possible touchdown of a “large” tornado in Joplin. Prior to my arrival in Pittsburg (which is about 25 minutes away from Joplin as the crow flies), I was a volunteer at our county fire department back in my hometown. Such a role breeds instincts that are not easily broken. My immediate response to news of the tornado was to want to get there as quickly as possible to help those in need. Unfortunately, recent surgery on my knee precluded any possibility of such action. But that didn’t matter. I had to do something. And I had other tools.

That’s where social media came into play. I plugged in, set up filters, turned my network on its head, and burnt some saved up social capital. As @ErikSTL noted, I sacrificed a bit of sleep in the process. My apologies to those that put up with my constant barrage of information during that time. So that’s the setup. Days of non-stop Twitter, Facebook, and Google Doc usage has left me with some stuff to share, and how it’s relevant to higher ed.

#1: It’s Called “Crisis” Communication For a Reason

Look, if a disaster hits your campus, a real disaster, be prepared to throw your best laid plans to the wind. And more importantly, understand that there will be a lot of people working damn hard to do the best they can in the middle of a lot of pressure and chaos. I saw a lot of complaints early on about the lack of this or confusion about that. That’s going to happen no matter what. We plan, we train, we drill, but really nothing is going to catch all the things that can and will go wrong in an emergency. Expect failures. Cell phone towers gone or overwhelmed. Phone lines down. Servers destroyed. How many of you have communications plans that rely on the assumption that communication channels will work as expected? Take those away and what will you be left with? A crisis.

#2: It’s Okay to Sacrifice Some Accuracy For Speed

There is a balance here. I’m not saying to just put out whatever random garbage you hear on Twitter. Ultimately, you never hear people being criticized for trying to push out too much information when it’s in good faith. Odds are, some of what you say will be wrong anyway despite your best efforts. But in an emergency, people on the outside crave anything that can help them be informed. You can easily specify if something is unconfirmed or not. For example, I heard an early estimate that had the death toll at 160, before it was officially even at 100. Later, people were saying 1500 were missing. Actual number: 232. Number of people criticized for pushing the 1500 number? 0. If you send out something like that, follow up after the fact with official data to correct the earlier inaccuracy. Tell people you’re researching unconfirmed reports. That serves to protect against rumors and also comfort people that you are actively seeking information for them. You can almost always issue a correction, and that’s okay.

#3: Who Do You Trust?

In your plan (for what it’s worth), identify someone that can be a voice for the organization. It doesn’t have to be some big, official person. It needs to be someone that will care, and will be safely isolated (physically) from the events so that they can do the job and hopefully avoid the disruptions that will occur within communication infrastructure. This person will help find answers for people, and direct folks to the proper resources. They serve as the traffic cop. People want a single, trusted source for information if at all possible. More than likely, whomever runs your social media now would be ideal for this role since they already know how to tune into the channels and talk to people.

And if that scares you, while you’re debating who should be doing it during a crisis, the audience will have already gone off and started listening to a dozen different, less accurate, unofficial, but more responsive people.  It’s not about having the perfect person, it’s about having a valuable person.

#4: Get Ready For Pain

In a crisis, something that is major, be prepared for hard stuff. What follows is the hardest tweet I have ever had to reply to, to a gentleman looking for his mother.

Joplin2 Lessons Learned: Joplin Crisis Communication

It took me about ten minutes to put together a response to his message. Why? 26th and Jackson was one of the hardest hit areas. There was, quite literally, nothing left there but rubble. In a crisis though, people will reach out wherever they can for hope, and that person might be you. Be there for them, and offer it, even if you know it’s going to end badly. And this stretches beyond the immediate event. After the tornado, dozens were searching for the missing. In many cases, people were reunited through the help of Twitter, Facebook, and those on the ground. Other times, they weren’t. I would be lying if I said it didn’t get to me after a while. But if you are prepared for it, you can do a lot of good and help folks before it overwhelms you.

Mr. Bell’s mother passed away that night due to injuries sustained during the tornado. I still think about it.

#5: Know Your Tools

Joplin3 300x172 Lessons Learned: Joplin Crisis Communication

Custom Google Map of Tornado Info

Twitter, Facebook - These are obvious. But understand there are other resources available as well to help communicate in a crisis, especially with respect to preparing information once communication is restored in an area if it has been disrupted. For instance, by the morning after the tornado, I was running the largest single list of donation locations using a crowdsourced Google Doc spreadsheet. In under a week the list grew from 25 starting locations to over 200 all over the country. We worked similarly to create a list of missing persons with images, last locations, who to contact, etc until the state was able to coordinate it’s effort. Others still put together a with a number of information points.

But there’s more than that. Have you heard of Sweeper? It’s an open source tool you can use to bring in multiple sources of data and get it structured in a way that lets you normalize the information. By itself, it’s neat, but combine it with something like CrowdMap (a hosted version of the Ushahidi project), and you can begin to create a powerful, real-time, geolocation based tool. There are also services like the FEMA National Donations Management Network system that helps make sure the right resources go to the right places, and the Red Cross’s Safe and Well project, and the Google Person Finder. Get familiar with all this information, as well as local and state tools that might be made available during a crisis.

#6: Listen, Provide, Anticipate

Set up your Twitter searches, monitor Facebook pages, turn on Google alerts. Get your ears open. This is for two reasons. One, you need to know what people are looking and asking for, but secondly, you need to make sure you aren’t either missing important information or lacking it somewhere. Information can and will flow two ways, and don’t assume you know everything about what’s going on around you. Odds are, you don’t. Once you have filtered out the noise from the signal, it will allow you to respond and push out important information, as well as plan ahead. For instance, the Red Cross knows from experience that in times of disaster, people want to donate, but tend to just use it as an excuse to dump old clothing. The result? Tons and tons of useless clothes that still have to get sorted, stored, and disposed of. That all produces a lot of extra work and overhead that gets in the way of real needs. Knowing this, before donations start flowing in, they can try and discourage the donating of clothing to save that burden. By being cognizant of such things, you can keep the system moving forward smoothly, not having to pause, backtrack, and rebuild momentum each time something doesn’t go as planned.

#7: Traffic Models Change

Everything you know about where people come to your site from is useless. The entire model you have is busted when an emergency occurs. Jonathan Steffens over at MO.gov put some information together in the following slides (21-24) about how their traffic was impacted by the disaster in Joplin.

Note that even within the same week, the social media profile flipped on its head as the crisis progressed. Early on, it’s all about Twitter (instant info), later people shifted to Facebook (static, sticky info). This is important as it should impact how and where you share information. It should also go without saying that you better have pretested your hardware to make sure you can handle any unexpected surge in traffic due to an emergency. This is one of the few time that really paying attention to your analytics in real time can be very valuable.

Plan Anyway

Even though you can be pretty confident that any disaster plan you make will start breaking down the second it is invoked, it is still important to have a plan. Even stripped down, it is going to serve to provide an important foundation until more permanent or official things can happen in the wake of events. And that plan should extend beyond just you. For example, MSSU was on the opposite side of town from the tornado, but they were quite obviously impacted since they had the space and resources to play an important role in the relief efforts. Our schools are play major roles in our communities, so think beyond your walls.

Conclusions

There you have it. Personally, I think one of the most important things you can do, if you can’t be on the ground, is help communicate. Like manual labor, it’s important to help, but not get in the way. Don’t share out information that is purely speculative. Try to get data from official sources as much as possible. Never underestimate the power of the voice in situations like this. It can be calming in a sea of turbulence when someone can turn to you for help or advice. Personally, I think that was one of the biggest failings through the event - that communication lacked where it probably shouldn’t have, crisis or not. For instance, FEMA should have someone on call to do what many of us did, just serve as a point person until more official wheels can find purchase. With a single phone call they could get someone online and talking within minutes. Why hasn’t that happened?

Technology is both a huge asset, but costly taskmaster at the same time. It can do great things, but it also can’t get out of its own way. At times, it was simply impossible to really follow the Twitter stream for Joplin. It moved too fast. At that point, it becomes useless. But it doesn’t have to be that way. Tools are within our grasp, and I think before long we will be doing amazing things in times of emergency. If you are looking for ways to help in the field of crisis communications, I STRONGLY encourage you to get involved with something like the Ushahidi project. While good, I see a lot of untapped potential there, and they are running mostly on grant funding and could use some good community resources to develop with.


Photo Credit: cc icon attribution small Lessons Learned: Joplin Crisis Communicationcc icon noderivs small Lessons Learned: Joplin Crisis Communication Some rights reserved by twi$tbarbie (derivative by permission)


This post was written by:

Director of Web Marketing - who has written 80 posts on .eduGuru

Director of Web Marketingjoined Pittsburg State University in Pittsburg, KS (NOT Pennsylvania, they spell it wrong anyway) inand is currently the Director of Web Marketing.  He is also CTO for the interactive map provider nuCloud. Web development's role in interpersonal communication is a principle focus of his efforts to improve and enhance higher ed web commodities.  He is an active supporter of the dotCMS community, accessibility advocate, freelance consultant, frequent speaker at web events, and general purpose geek who wears many hats.  Read his complete bio.


  • Steve -->

    Thanks for the post.
    I work at a Canadian university’s emergency center, and always appreciate seeing real-world “lessons learned’ articles to pick up new tips and insights.
    Good stuff !

  • https://twitter.com/andrewcareaga Andrew Careaga -->

    Good analysis of communications during a crisis of an almost unthinkable scale. The amazing thing about this event, from a social media communications perspective, was how so many pitched in to help. As with any breaking news event, there’s some noise and lack of clarity in the channel at the outset. But as you point out, immediacy trumps accuracy during those chaotic early moments.