Post-alert messages should confirm safety first, not explain the error

Post-alert messaging is getting a lot of my attention these days. For those unfamiliar with the term "post-alert message," these are the equivalent of "all clear" messages. But instead of simply saying "all clear" or "orders lifted", a post-alert message is written in plain language with a focus on communicating safety. We started thinking about the importance of these messages when we noticed how they are filled with jargon that implies safety rather than actively communicating safety.

When we did the research for the post-alert lexicon , we were focused on communicating the end to an imminent threat, but we also had two types of false alerts in mind - the one that was issued due to human error (think, Hawaii), and the one that was issued as a system malfunction (think, earthquake early warning sent to an area where there is no earthquake). In Hawaii, people waited for 38 minutes to receive that safety message. In Ridgecrest, CA, people waiting even longer, primarily because they didn't understand the earthquake early warning system would issue a message due to earthquake detection, not prediction.

We learned from our interviews and survey research that what people need in a post-alert message is an assurance that the event is over (or never took place) and a statement that they are safe.

It turns out the Nation Fire Protection Association (NFPA) has thought of two other types of false alerts - those that were well intentioned but incorrect and those that are sent maliciously with an intent to harm. The well-intended alert that was incorrect is probably most similar in quality to the human error alert. People need to know that they are safe. The malicious alert raises all sorts of new questions, especially if this means the system has been hacked and there will now be questions about who has access and whether any message coming through that channel can be trusted.

Importantly, none of the post alert message templates that we proposed or tested ever addressed the role of responsibility or attribution. This is because the first question people have when they are told "you might die" is "is it over"? We certainly know that people will want to know who or what is to blame and what is going to be done to prevent these errors in the future, but that information is not included in the first message of safety.

Some organizations, like the USGS, have thought through and planned the type of post-alert message they will send for each type of false alert that is issued (due to errors, malfunctions, or any other undetermined reason). Identifying the steps they will take ahead of time helps to initiate a post alert message immediately following a false alert.

So, here is my question for alerting authorities - what is YOUR plan for issuing that first message after the false alert goes out? Have you discussed this with your team to determine under what conditions you will send a follow up message and what it will say? Have you determined the appropriate content for each type of false alert that may go out?

For imminent threats events that that have been resolved and the environment is safe, an evidence-based post-alert message looks like this:

Message Source: The [name of threat] we previously alerted you about in [location] has been resolved and is no longer threatening. You can end the [protective action we told you to take].

In my opinion, a message following a false alert should still emphasize safety. In fact, following the structure above will ensure people understand that the initial uncertainty has ended. It could also use plain language and clearly state that the initial alert was false. This removes any ambiguity about the context of how/why that message was issued in the first place. A second message issued through different channels, could address the steps that are being taken to prevent future false alerts from being issued.

Be sure to download the post-alert lexicon to see the templates we have devloped for 50 different hazards. And watch this space for an announcement of a future updated lexicon that will be released as part of the Warning Workbook.