The social intelligence conundrum

I had an interesting dialogue on Twitter over the weekend about the need for agencies/organizations to use pay services to garner situational awareness in a crisis.

Here's how it started:

1. Matthew Kaskavitch ‏@socialmatt   Aug 9

   Do you have access to a social media monitoring service? If not, you're definitely unprepared for digital crisis comms. #smem

   Expand
   

   •  Reply 
   •  Retweet 
   •  Favorite 
2. Patrice Cloutier ‏@patricecloutier   Aug 9

   @socialmatt you don't need a service for that ... anyone can monitor SM for free with right training/policy and integration into EM #smem

   Expand
   

   •  Reply 
   •  Delete 
   •  Favorite 
3. Matthew Kaskavitch ‏@socialmatt   Aug 9

   @patricecloutier I don't believe there is any way someone could be as efficient in terms of situational awareness without a service.

I particularly pointed out the contribution of digital volunteers, most of whom, use a variety of free tools to do a fabulous job during emergencies around the globe:

Patrice Cloutier ‏@patricecloutier   Aug 9

@socialmatt tell that to all digital vols like #VOST and @HumanityRoad .don't need paid service .i'd trust their monitoring chomps over any1

The key question really, is what's needed? What do EOC managers or incident commanders need? Do agencies need access to the full "twitter pipeline" ?

I make the following points:

1. with the proper search/monitoring procedures (keywords, hashtags, geo-fenced searches), agencies can get a pretty good sample of tweets/social posts to discern trends, identify rumours, etc .... often with the support of digital volunters (VOSTs come to mind)
2. In the current state of the adoption of #smem by emergency managers ... there is simply no demand to get the whole data set ... to most in the EM community ...it'd just be noise ... that's where digital volunteers come into play.

Data without analysis or meeting the need of the emergency management organization or agency is useless ... It's got to support effective decision-making and online engagement.

So, a representative sample (and that figure keeps increasing with every single disaster) is more than enough to conduct social media monitoring. Again. this can be done for one or more of the following five reasons: 

• validate emergency information put out by an agency (are people doing what you want them to do ? )
• detect rumours/misinformation that could pose a threat to public safety/public health
• identify and route through appropriate channels, calls for assistance (although you won't see them all ... you need a process in place to have it go to the right place)
• detect reputation threats that could impede the agency's ability to respond (like key influencers criticizing your response or calling out senior executives or elected officials)
• to enhance your situational awareness (gather social inter via posts, pictures, videos that people share when they witness an incident/disaster)

The fact is ALL OF THAT can be accomplished by using free tools (Hootsuite or Tweetdeck for example) although their paid version adds some umph ...as does a tool like @geofeedia.

More important than using paid tools though is the ability to validate data from your social listening operation. More and more of these tools that combine some analysis and verification features are now available. A few here:

• digital content verification during emergencies ... a great handbook
• the AIDR (artificial intelligence for disaster response)
• the Verily project ... which capitalizes on the growing importance of crowdsourcing info during disasters 

What do you think? Paid or free ? the tools or the training and motivation of a dedicated team of volunteers ? I chose the latter ...

Digital volunteers and the emergency management community: strange bedfellows or natural mates?

I recently produced a three-part series of posts on the integration of socially-convergent tools into emergency operations centres and incident command posts. This week, I had the occasion to take part in very interesting discussions on how to integrate the work of digital volunteers using social convergence with official efforts from EMOs and government agencies. 

The meeting was to prepare an experiment that will look at fostering that integration across Canada (and the US too ! ). It's part of the broader CAUSE III program. The experiment in Halifax (Nova Scotia) will also coincide with the Red Cross Disaster Management Conference. 

Now during the meeting this week, long discussions occurred around how best to manage the flow of data/info created by digital volunteers during emergencies. Whether it comes from global organizations/Volunteer Technical Communities or VTCs (Standby Task Force, DHN, Geeks Without Bounds, etc) or local Virtual Operations Support Teams (VOSTs) or even spontaneous collectives, that input needs to be part of the situational awareness equation in EOCs. 

While everyone agrees on the benefits (better info to support decision making and online engagement), the question of the place to give to the info provided by digital volunteers  in the EOC or command post remains.

I stated on many occasions that my belief (especially as a leader in CanVOST) is that VOSTs are well positioned to be the link between VTCs and the emergency management community, But, the VOST main mission is still to provide surge capability to monitor the social sphere during incidents. 


But the debate rages on: 

• How can agencies tap into collective/spontaneous virtual initiatives that spring up during any large-scale incident? 
• How can VTCs that most often work in supporting relief/response/recovery efforts in third world countries best work with governments in the first world? Will they? 
• Is it better to have a group (such as a VOST) made of people with experience in both the VTC world and the EM realm act as a go-between? 
• How can organizations built around diffused command and control integrate their operations (or output ...) with quasi-militaristic, top down set up like we see most often in EM? 

Part of the experiment will be to provide some possible answers to these questions as VOSTs (CanVOST and other teams in the US) will be involved, as will VTCs (notably CrisisCommons).

Here's a look at how it's happened so far.

Stay tuned ! 

Part 3: social convergence in the EOC/ICP ... Logistics and Fin/Admin

In this third and final post in this series on social convergence in the EOC/ICP, I'll address the benefits for the logistics and fin/admin people.

I first looked at Command (June 9)  and then for Ops and Planning (June 24).  I also wrote a guest blog post for Geofeedia on the role of social convergence (with a focus on social monitoring) for situational awareness and the efficient allocation of strategic resources. 

For logistics, as well as any other part of the disaster response/management enterprise, social convergence (mobile tech + social networks) offers one key thing: speed. The ability to grasp in almost real-time what the needs of the populations impacted by an emergency might be. 

This might start in the EOC itself with simple geo-fenced monitoring, coupled with hashtag searches for such things as #help, #supplies ... all linked with the incident-specific #. You can quickly match needs and offers of assistance.

Here's examples, I just came across: 

WenatcheeworldJul 18, 7:03pm via InstagramWant to volunteer or donate to the #NCW Fire Victims in #Pateros or other communities? We have an…instagram.com/p/qnDGaADkfm/

#WAwildfire victims need some help in #Brewster and#Pateros near the #CarltonComplex fire. Local businesses are... fb.me/3w9iUPZLL 

One of the most illustrative examples I saw followed the devastating tornado that hit Henryville in Indiana in 2012. Almost immediately, calls for help, and offers for assistance, emerged on Twitter. Local official made the best use of this. 


Here's what it could have looked like in their EOC's logistics section:

A simple two-column dashboard on Tweetdeck (one of my favourite monitoring tools) showing #henryville and #help on one side, and #henryvilleneed on the other. Are you ready for this kind of direct, almost immediate flood of requests and offers of assistance?

It's exactly this kind of problem (in my opinion, opportunity really ...) that the O'Neill sisters (Caitria and Morgan) solved by making the most of social convergence and creating Recovers.org. To quote an article from the Huffington Post: 

The site works by linking volunteers with where they're needed in specific communities, offering up-to-the-minute social media updates on relief programs, and databasing even the most random donation items so they're easier to disperse exactly where they're needed.

Social convergence is a community resilience booster! It works. It's done so in Boulder and many other communities. My good friend Kim Stephen (@kim26stephens) expanded on that fantastic example in a blog post.

During Hurricane Sandy and its aftermath, crowdsourcing supplies such as gas, the availability of medical equipment and facilities, became common. In other words, supply chain professionals don't have to do all the work alone in a crisis ... the crowd can (and does) help. A good article in this PDF on page 26. The trend is being studied and validated across the globe by academics and business schools.

But for many, it begins much closer to home ... with insurance companies being able to grasp the situation in real-time. They use GIS, social media, to target their response as efficiently as possible.

So, enough about logistics .... point made. 

But what about our fine fin/admin folks (I'm always nice to people who ensure I get paid ! ) ? 

Besides, using online tools to process data and store content, they can use a multitude of mobile apps specifically designed for emergency response. Again, it's about bringing the info directly into the hands of those who need it NOW.

So, if you're not there yet ... it's not too late to get "socially convergent" and bring your EOC and your incident command team into the 21st century. 

Post #2: social convergence in the EOC/ICP ...plans and ops

In the first post in this new series, I talked about how socially-convergent tools could be used by command staff and incident commanders. Today, I'll explore what the future might hold for the Operations and the Planning sections within an ICS/IMS structure.

But before, I get to this new chapter, I'd like to reiterate the power of mobile technologies in the hands of first responders, SAR/HUSAR teams and the such. It's not just about the use of social media in emergencies. Social convergence is much bigger than that!

From adding to a search dog's natural abilities, to receiving images coming from a bouncing ball full of cameras, creating 3-D maps on the go in destroyed buildings and making the work of rescuers easier, the power of mobile devices has barely been exploited:

It's clear to me that in the Operations Section, social convergence brings new, validated tools and procedures that can be used by a single resource ... up to the Section Chief.
From dynamic, real-time, GIS mapping showing up-to-date situation on the ground, to the use of drones to support efficient decision making, everything points to the integration of mobile technologies (phones, drones, satellites) to make response efforts more efficient. And all that can be held in your hand ... on a smartphone or tablet. 

Now, imagine combining the data from drones, crisis maps and crowdsourced information, in real-time, to give the response leader all he/she needs to deploy resources in the most strategic fashion possible. 

Want rapid damage assessment? What better than a cheap drone ... keep your staff out of the danger area. Here's what the aftermath of a recent tornado in Angus, Ontario looked like from a drone's eye view:

In the planning section, the obvious benefits of social convergence are in the creation of better maps. Maps that reflect reality as experienced by the people impacted by a disaster or emergency. We are squarely in an era of community-based situational awareness. An understanding of how an incident evolves based on the experiences on the ground and not simply analysis conducted in an EOC far away.

This might appear daunting for the Planning Section Chief or even the EOC manager/director to grasp. The fact is, there is help out there! More and more digital volunteers are stepping up to support emergency management agencies. Virtual Operations Support Teams (or VOSTs) are now active in many countries (including Canada). 

Furthermore, that work is constantly being validated by sound academic studies and projects that keep giving it growing legitimacy. The results are in: VOSTs and other digital volunteerism efforts (mainly crisis mapping) save lives and speed up response/recovery efforts when whole neighbourhoods are burning or when large storms affect an entire country.

As the Plans Section Chief, you can use socially-convergent tools. Your resources unit can combine check-in apps, mobile devices and mapping to lay out a clear picture of what's available at any time. Again, the power of mobile is only beginning to be discovered.

For the Situation Unit ... mapping as previously described is greatly enhanced by social convergence. But sharing sitreps is made so much easier by the prevalence of mobile devices (smartphones, phablets, tablets) often "ruggerized" for field use. With a bigger screen info can be consumed much more easily by responders because it's provided in a much more visual manner.

The Documentation Unit can use a variety of cloud-based apps (google docs, dropbox, others) and services, all accessible by mobile devices to do its job and support operations.

Even your Demob Unit can use mobile devices and services such as check-in apps, cloud-based spreadsheets to keep track and help it plan the post-op draw back. 

I'm only scratching the surface here .... in the next installment in this series, I'll look at socially-convergent tools for the Fin/Admin and Logistics section ...

Social media in the EOC or the incident CP ... a look at what could be.

I've written in the past about the need for organizations to operationalize social network listening during incidents. It's become an absolute imperative. But how do you actually do it? What tools can you use to make it an effective addition at the incident command post and/or the EOC? 

In the first one in a new series of posts, I'll take a look at how command staff can make the most of social convergence in emergency management.

It starts at the scene. Today, there are tools that would allow any response leader to know what's on the social sphere when showing up at an active shooter situation, a large urban fire or some industrial accident. 

Case in point, the San Mateo pipeline explosion in California a few years back. It took a while for firefighters to realize there was no plane crash but that the pipeline blew instead. A look at Twitter might have told them right away. Listen to about 50 seconds into this video.

A tools such as Geofeedia might be a good one to have on a tablet in first responders vehicles. Any incident commander (I/C) could have a quick look to see what's up on social networks around the site.




Lets take a look at what the other command positions could be doing with socially convergent tools as the incident grows bigger. 

The Liaison Officer (LO) could start a group on Twitter and send out info that way. Or the LO could use Yammer or a private Facebook page to ensure inclusive comms where info could flow.

The Safety Officer (SO) could use Foursquare or even Google+ as a way to keep track of personnel and share relevant info via mobile devices. The SO could use Pinterest to post notices and other safety-related items. Again, this shows tablets or smartphones are a must for responders of any kind now.

The Public Information Officer (PIO) can use of combination of social networks (Twitter, Facebook, Vine, Youtube,etc) to send out info as he and the I/C decide on what needs to be communicated. They could even harness the power of mobile alerting through tools such as Ping4 Alerts! and similar apps. The role of PIO is evolving fast!

The Intelligence Officer (IO)  (OK ... to me it's a command function ! ) could use a laptop or a powerful tablet to set up an incident-specific mobile social media listening dashboard. Hootsuite, Tweetdeck and a few others come to mind. They have the ability to track hashtags, do geo-fenced searches and much more. It's a definite priority in my mind.

All this to say that technology exists to start this almost as soon as any incident emerges. No need to wait for a full EOC set up or a full fledged social media command centre to be set up.

What needs to happen though is procedures to be written, training to be thought of and given ... but more importantly for minds to be opened ! 

Grappling with social engagement ... some hard learned lessons ...

It's been a hard week for some big organizations on social networks. Many are learning that engaging online is a double-edged sword. But are there ways to mitigate the risks? I propose to show a few.

First, Air Canada went through some difficulties when one of their baggage handlers was caught on camera throwing luggage around .... 

I put together a storify piece on this whole mess. But how did Air Canada do overall in their crisis comms response? The answer ... a fair job at best. 

What worked: 

• They engaged with media outlets fairly rapidly 
• They engaged one-on-one on Twitter and Facebook with people who complained 

What didn't work: 

• They took a full day to post at-large on their Twitter and Facebook page
• They left the public space unoccupied for a long time letting criticism and negative impressions go unchallenged in the largest social forums.

What they should have done: 

• Engage broadly on social networks right from the get go ... don't let the social team work in the dark ... they need to be integrated in the crisis comms team ! 
• Use soundcloud to get a senior exec express his/her regret to customers ... a couple of hours after engaging online and distribute on Twitter/FB
• Use Youtube to broadcast another, more fulsome apology and promise to improve/correct ... just like what Fedex did just a couple years ago: 

Second is a story of a campaign gone wrong. In Ontario, we have a cartel of foreign beer companies controlling beer sales ... The Beer Store. Pressure is now on to allow the sale of beer and wine in corner/convenience store (a shocker really .... this would cause the end of Ontario who'd become like Québec where such sales happen: cool, hip and fun !  ... but I digress. )

The problem is that whoever is advising the Beer Store made them overplay their hands with really obnoxious TV ads. Ads that border on equating store owners with predators ... Not reflective of the reality at all:

They also had a twitter hashtag #ONbeerfacts .... Well, let's just say that things backfired there as well: 

@ONBeerFacts aren't fooling anyone with their ridiculous ads. Scaremongering at it's finest. #ONBEERFACTS

Rene LaBonte @QuestionWhatUC

#ONbeerfacts. Thank you for promoting your extortion of Ontarians

Here's an interesting Storify on the whole deal. And the reaction is not any better on the Beer Store's Facebook page.

What could they have done instead?

• If you're going to engage online and intend to create a real debate ... don't engage in silly propaganda elsewhere ...undermines any credibility you might have.
• They should have better gauged the risks associated with promoting their position on Twitter
• bring insightful and believable spokespersons to the fore ... talk about communities ... don't talk down to them ....

Finally, the week was topped up by the #myNYPD fiasco.

What were they thinking? They must surely know that the NYPD's reputation has been tarnished since the whole Occupy movement took to the streets?

While wanting to engage online is a noble idea (and some think it wasn't all bad ....) ... you've got to know how to pick your battles. This is not about judging how police behave in general ... or a discussion on the validity of accusations of widespread police brutality ... rather it's a question of evaluating the public perception before launching such an online campaign.

To say the results are not what the NYPD expected is a gigantic understatement. The twitter stream was inundated with images like this one:

What they were after was more like this: 

What could have done better:? 

• Once they realized the campaign had gone "south" on them ... engage even more ... tell their own stories ... even better, have their allies (and they have many) tell the stories on the behalf
• That should have started with the Commissioner .... who remained largely silent.
• Do some basic searches about your audiences' perception of the NYPD ...basic sentiment analysis would have sufficed ... 
• Make sure your social network team liaises/works with the crisis comms team ... now a must

What they have done right? Promise to stay the course and keep engaging on social networks. That's great and i'm sure they'll learn from that particular lesson. 

In the meantime, it's been a huge train wreck that has caused other police services from around the US and oversees as well, to become embroiled in the social crisis ...

What a week ! 

Social media listening: shared imperative for emergency management and national security

I had the pleasure to be a panelist at a very interesting conference/workshop last week organized by the Canadian Forces College. The focus was on social media and national security.

I provided a short presentation on social convergence it deals with national security in its broadest definition: emergency and crisis management. This is in recognition that the highest duty for any government is to ensure the well being of its citizens. 



Here are some observations I made during the two day meeting: 

• Agencies (law enforcement, counter-terrorism bodies, military) are paying great attention to open source intel coming from social networks.
• Most agencies recognize that relationships between individuals are no longer linear and in person ... social convergence (mobile + social) is both a source of empowerment for the people and an enabler (for good or bad).
• The same basic activities in terms of social media monitoring apply to both open source intel work and for emergency management and public health purposes: real time listening and analysis, collection and validation.
• After listening to the opinions of experts, I'm convinced that the most basic social listening tools we currently use (Hootsuite, Tweedeck, Geofeedia and others) are more than sufficient for EM purposes although they do not fill all needs for counter-terrorism or national security issues.
• There are some less known tools that can help validate social data.
• Despite the fact that the number of people who activate the GPS on their smartphone is quite small .... geo-location tools are supplemented by other algorithms/tools and can provide a large enough sample to get solid intel for situational awareness in emergencies.
• The work done by the likes of Patrick Meier (@patrickmeier), Project EPIC and others ... gives the #smem community more than solid "academic" backing to validate this whole new field of emergency management.
• Despite some negative examples (Reddit in the Boston Bombings for example) most military and intel organizations recognize the value of crowdsourced info ... despite the risks of the crowd getting it wrong .... the potential benefits cannot be ignored. It's something emergency management has already put in practice.
• And, surprisingly, more than a few people at the workshop think Edward Snowden did democracy a whole lot of good by forcing a public debate about the power of the state in the era of social convergence ... 

Training the modern-day PIO ... what should he/she know?

There is no doubt that social convergence is putting extraordinary pressure on today's Public Information Officers (PIO) ... also called Emergency Information Officers (EIO)...btw: i prefer that term since no one should mistake the job for a PR gig ...). So what should an aspiring PIO be taught? 

I believe the curriculum (and this applies for one-day crash seminars to full term courses) should be divided into three main tracks: 

1. learning the basics of emergency management ...IMS/ICS ... where the PIO fits in the big picture and his/her role 
2. Crisis/Risk Communications: basic techniques to ensure effective messaging in various situations 
3. social networks and related technologies: how to use social media ... on what platform ... and, very importantly, how to monitor it during emergencies.

Lets first look at the EM requirements. You'd be surprised to find out how many people assigned to the role of PIO or EIO ... or to speak/coordinate emergency info for their agencies during crises/disasters ... have NO emergency management knowledge whatsoever.  So, what to learn?  

Start with the basics:

• basic IMS/ICS structure and the role of PIO within it
• how the evolution toward joint/unified command is impacting the PIO role 
• how to run a Joint Info Centre (JIC) or emergency info centre (EIC) and how to manage larger public affairs coordination.

Secondly, the PIO cannot effectively coordinate the planning/delivery of emergency information to the public if he/she doesn't have a solid understanding of how people access and process info during a crisis and how to communicate effectively. Hence the need to learn to use a crisis communications approach. 

This includes matching crisis comms planning with an organization's hazard identification and risk analysis effort to ensure that what concerns the public (and therefore communicators) is not forgotten by the EM or BCP/COOP specialists. What does that mean? It means focusing on the five Ps : 

• procedures
• people
• preparation 
• practice 
• platform

Finally, social networks and related technology is the big force behind the changes in the PIO world and in crisis communications. From the evolving tech in the EOC, and in the field (as used by first responders) to innovative tech in the media ...It should all matter to the PIO especially how mobile devices keep evolving.

At the most basic level, it means getting familiar with the most popular social networking platforms.

It also means keeping up with alerting and notification tools (most of which now include social networks and mobile platforms) and crowdsourced platforms that allow for a better comprehension of how an incident evolves (crisis mapping, validation tools, etc). That's where the social monitoring (or listening) component comes in.

Social media monitoring is now at the heart of the considerations that PIO face during incidents. From deciding who should do it ... to how to operationalize social data ... that element alone is enough to keep most PIOs busy for a long time.

So, what do you think that new PIOs/EIOs should be learning?