Friday, October 19, 2012

Big exercise = big lessons learned in SMEM

I had the opportunity this week to take part (mostly as observer/coach) in a big emergency management exercise. Trillium Resolve had more than 1,000 participants from 50+ organizations in a vast part of southwestern Ontario.

The scenario was based on a series of severe weather events impacting municipalities in four different counties and the largest nuclear power station in the world (with 8 reactors ...).

A scenario this ambitious proved to be too enticing for not trying to add on a social media component. As part of the design team, I promoted early on, a "closed loop" approach. This was done mainly to ensure that no exercise tweet or Facebook post would cause concerns.

The way I thought we could do this without overburdening the exercise players, was to simulate the "output" of a social media listening operation. Based on the "tweak the tweet" syntax, I developed a series of report such as this one. Whenever possible or relevant, I added geo-location data and pictures or videos to add to the realism. 

The original intent was to have the reports send via email to different EOCs and PIOs playing. What actually happened was that the exercise controllers, broke down the reports and included individual simulated posts/tweets on a fake news web site that had been created for the exercise. So it ended up that players, especially PIOs/comms people, actually had to monitor social media and web/media stories in real time. 

Now, if we had had the time and resources, we could have used a tool such as Simulation Deck which is rapidly growing popular among the military, academic and corporate sectors, as a tool to test social media monitoring capabilities.

By trying to inject some contemporary reality into the exercise play, we were trying to do four things in relations to monitoring social networks:

  1. develop our ability to keep up with the volume of data we needed to keep an eye on (social and traditional media)
  2. increase our ability to identify social network injects which posed reputational threats or we calls for info or help
  3. determine what data could be analysed and transformed into solid intel for decision-making purposes and putting in place the channels to flow that info
  4. finally, validate the process for engagement/responses
Since I wasn't playing, I wanted to test my own skills ... so I started a crowdmap of not only the play injects but also the real-life media stories and posts about the exercise ... It was hard to keep up. I've written before that social media monitoring can be quite challenging and requires a team effort.

So, in closing, some key lessons I learned (or learned again !) during the week. To be effective, a social media listening/monitoring program needs the right resources ... it's a big job! You need (in my estimation):
  • to do the basic listening and searches, 
  • to tag/analyse/collate/curate the info and make it into actionable intel or tasks 
  • to conduct the engagement piece and respond to calls for help, request for info or dispel rumours.
It's more than one person could do ... or even three. Also, another key piece is to put in place, ahead of time, the channels through which the output of your social listening program will flow.

And very importantly, find a "champion" in your organization's leadership, who believes in the validity of the info that results and acts upon it ... proving that social media monitoring is now an essential part of any response to a disaster or crisis.

No comments:

Post a Comment