Effective Emergency Communication, Hurricane Sandy, and xMatters (11/14/2012)

In the aftermath of Hurricane Sandy (aka Superstorm Sandy), we received a lot of queries from clients, analysts, and media about our business continuity & emergency communications services.  There was a wide range of information that people were interested in, everything from "How many messages did you send out?" to "Do you have any emergency communication stories that you can share" and now a lot of "Any lessons learned that you can share with the rest of the BCM community?".

There are lots of blog posts that I could write on any of those topics but to start the conversation, this one captures a lot of summarized information in one place.

Usage profile & analysis of activity

Since the arrival of the storm in the US North East was not a complete surprise (compared to an earthquake, for example), we saw message volumes start to climb the week of Oct-22.  Interestingly that dropped a little over the weekend before the storm, and then surged up as the forecast for where the storm was going to hit tightened up.  The actual "live" period of the storm and the immediate aftermath represented one of the largest surges we've ever seen in the usage of our system. 

Based on our analysis of how our services were used during the disaster, here's a brief timeline of what was going on:

Warning phase (Oct 22-27):

• Overall system usage and alerts delivered increased by 200%
• Messaging channel mix: 55% Voice, 30% email, 10% SMS, 5% everything else (pagers, smartphone push, etc.)
• We classified this as the "heads up" activity as the storm was building and the the media was starting to talk about the possibility of merging with a Nor' Easter and the term Frankenstorm started to get used
• A lot of the communication was making sure that people had the right information in hand to prepare for the storm and know what to do
• Significant portion of the messages were also validating that key personnel could be reached on a variety of communication channels should worst case scenarios come to pass

Active Storm phase (Oct 28-30):

• Overall system usage triples from the week before to 400% of "normal event" alerting
• Messaging channel mix: 50% email, 30% SMS, 15% voice, 5% everything else
• Surge of collaboration workflow messaging such as conference calls. At its peak, there were 5000 people going in/out of conference calls.  This represents business continuity teams and emergency responders coordinating activities during the peak of the storm.
• Status updates and last minute adjustments were a big part of the messaging volume during this period
• Historically, and again during this storm, we do see email surge up to the #1 communication channel as BCM plans, recovery documents, restoration service updates, etc. need to get disseminated.
• During the height of the storm, no one really wants to spend the 30-60 seconds to listen to a voice alert so rapid information delivery & response collection drives up SMS usage along with email.

Storm aftermath (Oct 30 – Nov 2):

• Overall system usage surges again from the previous period.  Now at 1200% of "normal event" alerting volume.
• Messaging channel mix: 75% voice, 10% SMS, 10% email, 5% everything else
• Finding out if people are safe with voice alerts and soliciting responses to confirm how they're doing is key at this phase.  Providing them a number to call into to retrieve alerts is also very important, and that was reflected in the messaging channel mix as well.
• Due to the size, impact, and scope of the storm, the rate of conference calls only dropped by 25% from the previous period
• This phase also tends to be document delivery heavy (usually via email attachments) as alternate work locations, remote work instructions, etc. get distributed.

Customer stories

These are probably the most requested following any business continuity event, and also the trickiest to share.  In order to protect their identities, I've changed various details, but here are some anecdotes you might interesting from 3 of our largest clients who relied on xMatters to keep operating during and the aftermath of Hurricane Sandy:

1. European headquartered company wanted to make sure that their 5000+ NY region employees received alerts well in advance with recommendations on preparation both for work and at home.  During the height of and following landing of Superstorm Sandy, they made sure that their employees were safe while coordinating resources to minimize impact to their business operations.
2. Software company with global staff and operations used xMatters to coordinate move of key functions to locations outside of Superstorm Sandy's reach ahead of the storm.  While the storm was impacting their east coast facilities, they were able to continue to operate and also able to keep close tabs on their employees in the region to ensure their safety.  Following the storm they used xMatters to marshal resources to help impacted staff with up to date information and location of essentials such as shelter, gasoline, and medical services.
3. NY based company used xMatters to send out updates early and often as Superstorm Sandy approached the core operations of their business.  This included the ability to disseminate plans that could be accessed offline, as well as enabling communication plans that allowed for small group control & initiation.  Ability to decentralize their BCM alerts as the full impact of the storm became apparent was key to their recovery efforts. 

Lessons learned and rolled into Best Practices

A lot of these are worth of their own blog posts (stay tuned!) but here are some highlights:

1. Can never have too many communication channels to connect with employees, customers, and business continuity teams
2. Rapidly assembling teams for collaboration can reduce impact and back-and-forth communication chatter in the middle of a crisis
3. More mobile.  The more you can do on a smartphone / tablet, the better.
4. Balance need for centralized communications vs. distributed to get the job done.  Best performing BCM teams during Sandy were those that had BCM core teams running the big stuff while designated leaders in the field coordinated local communication activities.
5. Big impact = long lasting effect.  Sounds obvious but judging from our service usage data, we're setting new baseline usage numbers after Sandy.

If you have any questions about xMatters business continuity communications solutions in general, or a specific Hurricane Sandy question, drop us a note at info@xmatters.com or comment to this post.

xMatters Reference

Blog- Originally created by Abbas Haider Ali.