Skagit Bridge Collapse Public Information: Media and Coordination

Skagit Bridge Collapse Personal Lessons Learned
Marcus Deyerin
Northwest Washington Incident Management Team

Trooper Mark Francis, the Washington State Patrol PIO for this region, tweeted out my phone number for media at 8:38pm. The first incoming media call occurred at 8:40pm, and then it was then was non-stop until about 10:30pm. This is reflected in the drop-off in my tweets during this period. Again, if I had utilized both of my phones, I probably could have worked a few more tweets in during that period.

Common media questions that were (and will always be) difficult to answer:

Q: What was/is the condition of the patients/victims?
As someone who works in the public health field (my day job is an emergency manager for a local health department) – I am acutely sensitive to HIPAA restrictions on the release of patient/victim information. In that way, it puts me at a disadvantage because I can’t claim ignorance in the same way another PIO might be able too if they over-shared patient information. However, I can’t honestly say I know what the bright shiny line is in regards to protected patient specific information in a crisis situation. In the absence of that precise knowledge (although I’ll be doing some homework), I try to look at it from the perspective of a family member. If I were at home, and I recognized one of those vehicles in the water as belonging to my spouse, sibling, child, etc. – would I really want to hear anything but good news from some random guy on the TV? I’ll never forget a colleague who worked in disaster mortuary affairs that once said, “Anderson Cooper has done more death notifications than any medical examiner in the country.” And he didn’t mean that in a good way.

Personal lesson learned: if I can share good news in a generic way, I will (e.g. “the driver of vehicle X was walking on their own – but otherwise I’m not familiar with [or can’t disclose] the extent of any injuries they may have sustained.”). But I will never speculate about someone’s medical condition when it’s unclear, and whenever possible, bad news about any specific individual should be delivered by medical professionals directly to a victims family. I know that’s not always realistic – but I personally think it’s a goal we in the public information field should aspire towards.

Q: How many “X”…
The media seems to have a fixation with numbers. I’m not certain why that is – perhaps it’s some kind of recognizable yardstick by which to compare one disaster with another. I personally find it a little annoying and unhelpful in the immediate context, unless it has a direct bearing on the public’s safety. But it is what it is, and it isn’t likely to change.

Personal lesson learned:

  • Be prepared to offer numbers for something – anything; but emphasize these are estimates and likely to change (because they almost always do).
  • Know the difference between estimating something and speculating (i.e. guessing). This distinction will be different for every individual based on your personal knowledge base and experience. For example, I felt comfortable estimating the depth of the water where the bridge collapsed based on what I saw; but for me to make any statements about the extent of someone’s injuries, or what caused the bridge to collapse, would have been pure speculation. Consider the difference between “I saw these two people walking on their own” [factually accurate] vs. “I saw them walking, so I think their injuries aren’t that bad.”

While the two questions above were the most common, there were many more I received that I simply wasn’t in a position to answer, either because I didn’t know (common), or because I just wasn’t the appropriate person to answer, or both. When that happens, all you can do is say “I don’t know but I’ll try to find out, or direct you to someone who may know.” This is where interagency coordination comes in (more on that below).

I only had one frustrating experience with media…and that was boundaries on time. Once the phone started ringing, it didn’t stop for over two hours. I actually think it would have been longer had there been more severe injuries / fatalities, or if the incident had occurred earlier in the day. Ninety-five percent of the reporters I was able to connect with respected that I had a limited amount of time to offer them on the phone. But there were two who stood out as remarkable exceptions. One asked the same three questions, over and over. It was almost as if she was expecting a different answer each time, or just trying to keep me on the phone for the sake of it. Another was obviously hand-writing every single word I said verbatim (or typing really poorly), and kept asking me to repeat what I’d said so she could catch up. I really wanted to suggest she learn short-hand, or buy herself a recorder; but I bit my tongue knowing I was making my own mistakes as well.

Now the good news. The media I engaged with were very professional. There were a number of times when they had every right to be frustrated themselves.When the incident commander established a restricted area around the bridge for safety reasons as well as to ensure responders had the space needed to do their jobs, the media respected that, even though it took away their best visual vantage point of the collapsed bridge. I was able to get the IC to open up a portion of the river bank for media to get their camera shots (but still not accessible to the general public); but even if the IC hadn’t agreed to that, while they may not have been happy about it – I’m confident the majority still would have respected the boundaries without cajoling from law enforcement.

Later that evening, there was mixed messages sent out about where the Governor would be doing his initial press briefing. We had media staged on both sides of the river, and of course the Governor’s team came to the opposite side of where the media were expecting him (not intentionally – it just happened that way). Because there were some media there, that’s where he conducted his first on-camera statements. But once the Governor’s team learned about the confusion regarding the briefing area, they also did their part and travelled to the other side of the river to conduct a second briefing in the more formal location that had been prepped for that purpose.

I know these things sounds fairly trivial – but in the moment I was expecting at least one media type to express some frustration, but it didn’t happen. Again, the media I encountered were professional, polite and understanding of the challenges I and the other agency PIOs were dealing with.

Agency PIO Coordination
In a multi-agency response, the establishment of a Joint Information Center (JIC) is an imperative, and by Friday morning that element was in place. But in the early hours of this incident, a big challenge I encountered was the inability to coordinate in the field with the other PIO or agency reps I needed to. Had the phones been working, this would have been much easier. But even then, there were a lot of moving parts at the scene, spread out far enough (and in some cases separated by a fairly substantial river) to render the coordination and sharing of rapidly evolving information difficult. The best example regards the truck and driver that Washington State Patrol were investigating as having a possible (now confirmed) link to the bridge collapse. Media helicopters could clearly see WSP troopers converged around this vehicle, and they were asking me what I knew about it – which was nothing. By the time the media on scene were done with me and I had a chance to re-connect with my WSP counterparts, there wasn’t much else I could share with the media regarding the truck that they didn’t already know at that point.

Even if the phones had been working, I don’t think I was really prepared with a strategy to connect in the field with the number of agency reps I would have liked to spread out over a 1/2 mile radius (or 2 miles if you count the physical location of the EOC and JIC). From a personal lesson learned standpoint – I don’t have one other than the nature of the challenge itself. It might be tempting to say it’s a technology problem (the phones were down); but it’s an organizational challenge too. At one point I’d estimate there were at least 300 responder types on scene (fire, police, search and rescue, WSDOT, Coast Guard, even federal customs and border protection). That’s a lot of agencies to try and gather information from as the on-scene PIO. I’m still working on figuring out a good strategy and solution to address this (of which the VOST concept may play a part), and I’m certainly open to suggestions.


Tomorrow wraps up our coverage of the Skagit Bridge Collapse Public Information response by Marcus Deyerin. We’ll be talking some overview, some wrap-up, some mistakes and the most important part of any response.


Skagit Bridge Collapse Public Information: Social Media

Skagit Bridge Collapse Personal Lessons Learned
Marcus Deyerin
Northwest Washington Incident Management Team

Twitter, by far, was the most valuable tool for me to provide information to the public and media. Although I lost my ability to make calls and send SMS messages, I was able to send Twitter messages throughout. However, I was not able to successfully send tweets with a photo attached in those first two hours or so. After my second unsuccessful attempt to send a tweet with a photo, I gave up and decided to focus on text only tweets, since I figured the news helicopters were providing ample visual coverage of the scene.

It was only the next day that I was able to survey my “@ mentions” where I saw several media who were trying to contact me directly. They wanted me to either call them for an interview (which obviously I couldn’t during the first couple of hours), or to follow them on Twitter so they could direct message me. For a single PIO, this is a conundrum, because you want to be available to the media and Twitter isn’t a bad platform for that, but trying to monitor the huge volume of traffic directed toward you in those early moments is close to impossible. This is where a virtual operation support team (VOST) could really be useful. I’ll talk more about how I could/should have utilized VOST in a section below.

A couple people have observed and questioned the wisdom in regards to me tweeting in an official capacity from my personal account – most notable among them Gerald Baron. I am in total agreement with Gerald that this was not ideal, and it’s not something I would want to do again. Here’s why it happened… the Twitter account (@NWIMT) for the regional incident management team I’m on is what I should have used. Unfortunately, my day-job home agency transitioned to a new email system last week, which subsequently required reconfiguring my mobile device (which wiped my Twitter account info). That occurred on Tuesday – and I simply hadn’t re-added the NWIMT account to my phone’s Twitter client. When I attempted to re-add it at the scene, I couldn’t remember the account password, so I just had to go with what I had – which was my personal Twitter account.

I offer the above not as an excuse, but rather explanation. The obvious lesson here is if you rely on a particularly critical tool for something, you can’t wait even a “few days” to get it back in place. Having said that, we all live in the real world, and that kind of thing is just going to happen. My personal account is the tool I had available to me, and again – while not ideal – it served the purpose of getting the information out during the critical period when timeliness was everything. In a crisis situation, flexibility and adaptability are key; and good now is better than perfect later.

Personal lessons learned:

  • Twitter reigned as the superior tool for getting information out rapidly to a broad audience. [Note to Twitter – please, please don’t do anything vis-a-vis your API or business model to mess this up for those of us in the emergency management field.]
  • Twitter worked when phone and SMS didn’t. That won’t be true in every situation, but it was interesting nevertheless.
  • Photos attached to tweets are great – but may not always work in a constrained data flow environment
  • Once the media calls started coming in, I was no longer able to tweet. If I need to do this again, I’ll direct media calls to a different phone I have, so I can take calls on one phone and use the other phone for tweets / social media
  • If you’re sending tweets with time-sensitive info, add your own time stamp (e.g. 1015hrs). I remembered the value of this about half-way through my own efforts
  • The public doesn’t care about “official” titles – they value the quality of the information being provided. That’s not to suggest we in official roles shouldn’t care about which account we use; but we do need to understand that the audience will go where the best information is coming from – so if you want to be the official and best source – then you better be providing the best information.

I had a couple of emergency management colleagues contact me through Twitter to see if there was anything they could do to help from afar. I want to extend my appreciation to those folks (you know who you are).

This concept of “digital support” enabled by technology is increasingly being utilized around the county and the world, and is known as a Virtual Operations Support Team. It’s already been explained and highlighted elsewhere by others far better than I ever could [<– Jim – maybe link to other blog posts about this?]. When that support was offered to me, I didn't have the wherewithal to know what kind of assistance to request. But now with the benefit of retrospect, here's what I should have asked for:

  • Monitor my Twitter @ mentions for media contact requests. I simply didn’t have time to check @ mentions, and then subsequently filter out media requests / questions from the overall stream of retweets. The VOST could compile and forward the relevant ones to me via email, which I could then either respond to directly, or forward to the appropriate agency specific PIO.
  • Monitor overall social media and traditional media coverage. What questions do people have that aren’t being effectively answered? Are there rumors or mis-information we need to address?
  • Establish and populate an incident specific website. In a multi-agency, multi-jurisdictional response, this is perhaps the single biggest need during the early hours of the incident, but also the biggest challenge. It’s easy enough to populate the content quickly, but you have to have an existing website place (or dark site ready to go), and on a server robust enough to handle the surge of traffic you’re likely to receive.
  • Help build a media cheat-sheet. On Friday morning my second task was to put together an Agency Point of Contact sheet for the media. I noticed a lot of reporters were asking the right questions of the wrong people, if for no other reason than because they weren’t entirely sure which agency had purview over a given subject. I listed every stakeholder agency involved, the POC and contact info, and the topics or issues for which that agency was the most qualified to answer. After the 12:30pm briefing on Friday, I handed this out to media and they seemed to really appreciate it. It would have been really helpful to have this at least started the night before.
  • I know there’s more a VOST could have probably helped with, but these are the things that stick out in my rearview mirror. While there are a number of excellent established VOST organizations out there, this is definitely a capability we’ll be looking to develop further within the IMT I serve on.


    Tomorrow we’ll have Marcus back for an overview of his dealing with the media and coordinating with other agency PIOs. Stay tuned!

Skagit Bridge Collapse Public Information: Initial Response

Skagit Bridge Collapse Personal Lessons Learned
Marcus Deyerin
Northwest Washington Incident Management Team

Initial Response
A few people have asked me how I was possibly on-scene so quickly. Pure coincidence. My son participates in an athletic activity about a mile from the bridge collapse scene. I heard one, then two, then multiple sirens – and you don’t have to be an emergency manager type for that to get your attention. I opened up a radio scanner app I have, and the first words I heard were “I-5 bridge collapse over the Skagit River…”. I didn’t need to hear it again for confirmation – the number of sirens in the air was confirmation enough. I immediately grabbed my son and we headed to the scene. I knew exactly how to get there quickly since it’s a route I often run while my son is at his activity.

When I arrived, I quickly recognized the on-scene incident commander, the local fire chief and a former team member on the Northwest Washington Incident Management Team (NWIMT). I let him know I was there, but then just stepped back and stayed out of his way. After about 15-20 minutes I again approached the IC and asked if I could help in any way. That’s when he remembered my PIO role on the IMT and asked if I’d be willing to fulfill that function there on-scene.

With phones not working and the “field” nature of the scene, Twitter was the obvious and best platform for communicating information to the public. I spent about 3 minutes trying to get my team Twitter account functioning (more on that below), and then gave up and just started tweeting incident information from my personal account.

Phone and SMS
The phone system was impacted quickly (which I expected), but much more broadly than I would have anticipated. My colleague who was located up in Bellingham reported trouble making phone calls about the same time I lost my ability to call out. Here’s what my notes and phone logs reflect:

6:55pm – bridge collapses
7:10pm – I arrive on scene (estimated)
7:12pm – Outbound call successful (work phone)
7:30pm – Assigned as PIO by on-scene incident commander
7:31pm – First Twitter post sent from scene
7:37pm – Outbound SMS attempt – unsuccessful (personal phone)
7:48pm – Outbound call attempt – unsuccessful (personal phone)
8:09pm – Outbound SMS attempt – unsuccessful (work phone)

I’ve omitted a few redundant attempts from the timeline above for brevity’s sake – but you get the gist.

8:38pm – Washington State Patrol district PIO provides my phone number to media via tweet
8:40pm – First incoming media call
12:03am (Friday) – Last incoming media call before WSDOT took lead as incident PIO.

Personal lessons learned:

  • I don’t normally care for them, but in this situation I really wished I had a bluetooth earpiece for the phone.
  • We (emergency management) have been telling people for some time now that even when the phone lines are overwhelmed, SMS might still work. I think we need to emphasize the “might” element. In this instance, both SMS and phone connectivity started working again in the immediate area within 90 minutes or so. But that’s a long time if it’s your only way to communicate.


Tomorrow, Marcus will be talking about social media during the response, specifically Twitter and a bit about virtual support teams (VOST).

I5 Bridge Collapse Public Information, Part One

skagitI’ve had the distinct pleasure of being friends with Marcus Deyerin for a few years now. In just one of his jobs, Marcus works as the public information officer for the Northwest Washington Incident Management Team. If you followed the news last week, you’ll know that he had a pretty important job last week, especially after the 1-5 bridge over the Skagit River collapsed.

Marcus was assigned to be the PIO for the rescue portion of the operation and the first thing he did was post this on his Blackberry:

Just three minutes later:

And that began an amazing night of using social media to provide updates on a rapidly changing situation:

I’ll bet you’re wondering why didn’t he just release this information to the media and let them fulfill the role they usually do? Because of this:

Cell networks were down from everyone in the area being on their phones! Tell me you don’t foresee that happening in your emergency. As the cell networks calmed down, Marcus was able to get back on the phone and support the news organizations, but didn’t forget about the social media aspect:

He kept up updates for more than six hours that night. Through rescues and press briefings. This was a model social media operation.

What amazed me was how much the public was looking for this information. I noticed, on his first tweet that evening, that Marcus had 380 followers. By the next morning, after just an hour-long rescue operation, his followers had nearly doubled to nearly 700. Every one of his tweets was retweeted between five and ten times. And this was from a personal account!

I’ve asked Marcus if he could write something about his experience for posting this week, and depending on NWIMT, we should have a super series of posts. Keep an eye on this space for the latest!


Let me come right out and say it: I hate the term shelter-in-place. A lot of my emergency folks will shun me after saying it, but I can’t live a lie anymore. It’s one of those emergency terms that we use. Like the difference between warning and watch and advisory.

When I was younger (and more idealistic), I said that each time we have an emergency–even someone else’s emergency–we should take that opportunity to teach people our terms. Teachable moments, getting people into the emergency mindset and all that. What bullcrap.

How did we get so high-and-mighty and think that our terms were better than those used by the public? Why do we spill ink left, right and center trying to teach people our jargon? Because that’s what it is: jargon.

My biggest problem isn’t that people don’t understand what we’re talking about (though, admittedly, that’s a pretty big problem). My problem is that we continue to use our jargon, and then have to take time to translate it to normal people language. See Massachusetts Governor Deval Patrick speaking during the Boston bomber manhunt (fast-forward to :37 seconds for the interesting part):

He says, “We’re asking people to shelter-in-place. In other words, to stay indoors, with their doors locked.” Why does he say both? Because his emergency folks are telling him that people need to shelter-in-place, and his communication folks are telling him that people need to stay inside.

And we do it over and over and over. Here’s a potentially life-saving bit of advice. And now here it is again in plain language. Why not just say it in plain language the first time?

Actor Alan Alda has a great little project about plain language that I’ve been meaning to link to forever. He wants to get people to explain science in plain, regular language. His first attempt was trying to describe what fire was, to a eleven-year-old. Can you do it? Or will you default to your science jargon?

While the project is super-duper interesting (seriously, watch the winning video), this article highlights the danger of our continuing to ignore using plain language warnings:

I probably learned the best lesson about talking in plain words from my youngest grandson. We were on vacation in the Virgin Islands, walking on a path that led to the strangest tree we had ever seen. The trunk was covered with angry looking thorns. I thought, wow, this is a great chance to talk with Matteo about how this tree might have come to look like this. So, we sat on the ground and had a wonderful exchange of ideas about evolution for 45 minutes. He was only 6 or 7, but he was taking in everything I told him.

The next day he was swimming with his cousin and asked her a question about science. She said, “Why don’t you ask your Grandpa about that?” And Matteo said, “I’m not makin’ that mistake again.”

The worst thing that can happen isn’t that people get confused by our warning jargon. It’s that they will stop listening to us! Your shelter-in-place is enough to drive people to alternate, and probably wrong advice givers!

Who’s An Expert?

In those early morning hours, I dream that I am an expert. That I’ve established the kind of credibility that’s made my name synonymous with whatever weirdo topic I happen to be dreaming about. I’m famous and everyone knows who I am. I have no such illusions once the alarm clock goes off, as the scales are lifted from my eyes, though. That isn’t to say that I toil in obscurity. I’m very, very lucky that I get invited to travel a bit to talk about my experiences, and literally dozens of you will read this. In the end, this is better than a sharp stick in the eye, and I’m especially grateful for that.

But as much as I enjoy (and take pride in?) my humility, I know that I’m not the only one out there with some modicum of success, and I know that my humility is rare.

A quick search on any social network or website for “branding” will direct you to an orgy of bad advice and five-minute websites with low, low prices for consulting services. They are self-described, “ninjas,” and “gurus.” They can get you thousands of followers. And, given the time constraints that many of us in government communications have, some of these folks sound pretty good. A one-day consultation to get your social networks up and running? A templated crisis communications plan, just plug and chug?

I would urge caution, though. And not just because they’re my competition. But because their advice is probably not that good. That self-described, “Twitter master,” with 87 followers, probably isn’t. The twenty-two year old crisis communications wunderkind might not have the experience you’re looking for.

The reason I bring this up is because of two blog posts I saw earlier this week on exactly this topic. Geoff Livingston posted the excellent Differentiation Requires Show, Not Tell. His point was that people who tell you how wonderful they are and give lots of advice based upon best practices probably isn’t the best for your business. Folks who’ve done the work and actually used the advice (even when failing) they espouse are the ones that can help.

The Internet and in particular social media have empowered thousands, perhaps millions, to start their own businesses. One outcome of the social media movement is how easily people become “thought leaders” or topical influencers.

As a result, we have many paper tigers running about, almost indistinguishable from the ones with real teeth with one singular exception: Results.

Then Mashable followed up with advice on how to avoid those “paper tigers”:

“Rather like achieving academic tenure,” says Lieb, regarding one way to think about the process. “Thought leadership requires a continuum of wisdom, accomplishment, and a body of published work that stands the test of a degree of time.”

Same with Seth Godin — decades of proven concepts behind the notion that he’s an authority on the subjects that he tackles. So, the test of time and accomplishment is a big part of thing. The proof in the pudding matters. We are the curators. But it’s an ongoing responsibility. And every time a so-called thought leader self-nominates, we would do well to respond not by retweeting, but by saying something like: Hey, not so fast buddy.

Listen, social media isn’t the Wild West anymore. I’ve been publishing for nearly six-and-a-half years and I still consider myself a newbie. But my length of service isn’t intended to put you off. It’s intended to demonstrate that there are people out there who’ve been doing the work you want to learn about for literally years. There are best practices. There are ways to succeed. And that’s ultimately what you want right? For your agency to succeed in social media?

You don’t need some johnny-come-lately to give you insights; instead, look around for real-world experience and people who are passionate about what they do, not the millions they hope to make (and I assure you, there are those of us out there, especially in government communications, that LOVE this, and do it every day for FREE).

Collaborators, Not Targets

Last summer, we posted on audiences a few times. First on how your audience is a lie, then about how to communicate when there is no audience. The idea behind those posts is that an audience is a passive idea. Someone that just sits there and waits for your message. Thanks to social media, there are fewer and fewer folks that we are trying to message that act that way.

As much influence I think I have, those posts didn’t really change anything. We still talk about target audiences. We still write fact sheets that ignore 95% of the people that might read them. We still talk at people.

Sarah Larcker, writing for Marketing:Health, had a great post recently called, “I Am Not Your ‘Target’,” that made me think of those posts and how far we still need to go to properly understand our publics and effectively communicate with them.

When we generalize to “patients,” we lose resolution. We lose the person inside that patient. And “sufferers?” No one is defined solely by his or her relationship with a disease.

Most polarizing of all is “target.” When we call our customers “targets,” do really we mean that? Do we mean to aim our forces at them and barrage them with messages? Do we expect this to be effective in a world where they can so easily ignore us – and form their own opinions of us?

In emergency management, we’ve undergone a similar change in how we understand our audience as well. They used to be victims. Something terrible happened and we came in and saved the day. But the current FEMA Administrator, Craig Fugate, changed what he called them a few years back. They were no longer “victims,” they were now “survivors.” Dealt a blow, they’ve persevered. They are now excellently placed to help out now, to be partners.

Just that change in how we understand our customers has engendered a change in how we interact with them. We now look to them to volunteer, to help out on scenes. They are a huge part of emergency response today, and they weren’t before.

We, as communicators, need to make a similar change. We need to change how we think about our audiences. That starts with changing what we call them. They are not target audiences; they are not targets to be shot at.

Given that our messages are part of a conversation today–a conversation that is dominated by what they say, not by what we say–they are our partners, our friends. Our collaborators. If they fail to pick up our messaging, it is because we failed to collaborate with them. If our messaging succeeds, it is because they have taken the seed we started with and have amplified it to success.

Much like Dr. King’s voice may have spurred action, it was what his collaborators did with that message that changed the world.