Lessons Learned From Hawaii
How Emergency Alerts Should Work
This article originally appeared on Inside Higher Ed’s Call to Action Blog:
It was a wild weekend for our friends in Hawaii. On Saturday morning, more than a million people were sent a false alarm alert that a ballistic missile was headed their way. To say it caused extreme panic is an understatement. However, given my past experience as a public information officer with a designated seat on the university emergency operations team, I can empathize with the challenges the agency faced. We all can use this unfortunate incident to better understand how emergency operations work on the inside and to be better prepared as communicators.
Multiple communications methods (and debriefs) are essential.
The Hawaii Emergency Management Agency (HI-EMA) released an initial timeline of events the day of the incident, showing how quickly the situation evolved. Emergency notification systems often utilize multiple methods — outside sirens, text messages, emails, robocalls and social media, just to name a few. Typically, there’s a tiered order in how the system delivers the messages, and in my experience, email and text messages are usually the first to be received. Robocalls take the longest, as the system often has to wait several rings for a call to be picked up by an individual or voicemail. It’s recommended that the notification go out within the first 10 minutes of a crisis, but I’ve learned that it can take a considerable amount of time to send them out depending on the length of the list. It can be a few minutes or up to an hour, depending on where you are situated on the list. At California State University, Sacramento, our list was alphabetized, so I wouldn’t receive my alerts until they hit the S’s. I can remember one instance in which I didn’t receive any email or text alerts during a test at all. (Gasp!) But, as in Hawaii, we debriefed internally to find out why that happened, and as it turned out my cell phone number, which has been assigned to a former employee at the university, had accidentally been erased from the system after she left (the system routinely updates when students graduate or move on). Had that test not happened, I may not have known about a real crisis. That’s why the debriefs are important — they can mitigate future problems or gaps in the communications, as well as improve the speed at which messages are sent.
There are limitations, even with the most sophisticated systems.
When an emergency notification system is activated, it can tie up all the communication channels that the agency uses, which makes it even more difficult to cancel. After the initial alert in Hawaii was sent out at 8:07 a.m., the HI-EMA did dispel the threat and issue a cancellation within six minutes — a pretty good response time, in my opinion. However, they were hindered by the fact that they could only intercept the communications that hadn’t gone out, which still led to a lot of panicked people. And given that the second alert wasn’t sent to cellphones until 8:45 a.m., 38 minutes after the first alert, a lot of people were understandably upset. During that time they were trying other communications methods to reach those affected, such as social media, but it wasn’t soon enough. And unfortunately, no matter how hard you try, there will be those that never hear the alert at all. All you can do is try to reach as many people as you can.
Mistakes will happen, but we must learn from them.
Much like Hawaii, when we sent out emergency notifications (or even press releases) at the university, a screen would pop up asking if we were sure if we wanted to proceed, and you had to click it a second time to launch the system. But once that goes, you’re off and running. We had at least one time that an emergency test was sent without prior warning, and a few times we bungled and then corrected press releases. But much like this case, it’s always a learning opportunity to make things better. Hawaii made some immediate changes, such as making the activation/verification process for emergency alerts a two-person job and setting up an automatic trigger for a cancellation within seconds of an error. These were good lessons that a university emergency operations team can and should consider.
Test frequently, but judiciously.
Speaking of tests, we now know that HI-EMA had been testing its emergency alerts for a good month prior to the false alarm, given the tensions in North Korea. Many Hawaiians complained it was too frequent, and there is some validity to that argument. Sometimes the tests can be counterproductive — if people are receiving the notifications too often, they may tune out in a real crisis. But these tests are necessary, and can still be done with a smaller group of officials, even if the general public isn’t involved. In my former emergency ops role, I received test alerts that went out to a small group every two weeks. But if every two weeks doesn’t seem reasonable on your campus, testing on a monthly basis with a core group is another best practice. And for a test that goes out to the entire campus community, I’d recommend that those be administered one or two times a year. At Sacramento State we did an annual earthquake drill, as well as an annual emergency alert test. We would sometimes get complaints from students when we ran the public test, but more often than not the feedback we received was positive as they appreciated knowing emergency preparedness was a priority for our campus law enforcement. The Federal Emergency Management Association (FEMA) also recommends best practices for their Emergency Alert System, and EverBridge, a critical event management software solution for many higher education institutions, offers abest practices guide on their website as well.
Social media is a great way to communicate in an emergency, but not the only way.
Most state agencies and universities use Twitter and Facebook during a crisis and for good reason — you can deliver your message fast to your community of followers. What I don’t understand, however, is why it took 13 minutes — from 8:07 a.m. to 8:20 a.m. — to post the false alarm update on Twitter that said “NO missile threat to Hawaii.” (This was also 7 minutes after the cancellation was sent to the system.) If the emergency system tied up the social media channels during an activation, like it did at Sacramento State, perhaps the HI-EMA could have asked the governor or some other high profile official to tweet the update sooner on their behalf. That’s why it is important to always have a backup plan. HI-EMA was probably following their protocol in being the first to respond, but given the length of time, they should have been exploring other avenues to dispel the false alarm sooner. Perhaps relying more on their crisis communication phone tree (the emergency calls that often go from the top down) or calls to local media to verbally spread and post the message would have helped. During an emergency, all bets are off and you need to do what you can to get the information out quickly.
Lastly, what happened in Hawaii serves as a good reminder for us all that we need to be prepared for emergencies, not only at our jobs, but with our own families. My parents learned that the hard way on Oct. 8, 2017, when they were forced to evacuate from their home in Santa Rosa, Calif., at 3 a.m. due to the Tubbs Fire. The wildfire came within three blocks of their house, and thanks to a neighbor pounding on their front door, they were able to get out fast. They had not been previously signed up for emergency alerts on their cell phones, but they are now. You can check out Ready.gov for some good tips and what communications alerts to sign up for in your area. As the saying goes, “it’s better to be safe, than sorry.” In time, I hope most Hawaiians will agree.
Elisa Smith is a PR consultant with KP Public Affairs and a former journalist with close to two decades of experience working in communications for higher education, corporate and nonprofit organizations. She can be reached at www.elisabongiovannismith.com