How to Crowdsource Crisis Response

I recently had the distinct pleasure of giving this year’s keynote address at the Global Communications Forum (#RCcom on Twitter) organized by the Interna-tional Committee of the Red Cross (ICRC) in Geneva. The conversations that followed were thoroughly fruitful and enjoyable.

Like many other humanitarian organizations, the ICRC is thinking hard about how to manage the social media challenge. In 2010, this study carried out by the American Red Cross (ARC) found that the public increasingly expects humanitarian organizations to respond to pleas for help posted on social media platforms like Facebook, Twitter, etc. The question is, how in the world are humanitarian organizations supposed to handle this significant increase in “customer service” requests? Even during non-emergencies, ARC’s Facebook page receives a large number of comments on a daily basis many of which solicit replies. This figure escalates significantly during crises. So what to do?

The answer, in my opinion, requires some organizational change. Clearly, the dramatic rise in customer service requests posted on social media platforms cannot be managed through existing organizational structures and work flows. Moreover, the vast majority of posted requests don’t reflect life threatening situations. In other words, responses to many requests don’t require professional emergency responders. So humanitarian organizations should consider taking a two-pronged strategy to address the social media challenge. The first is to upgrade their “customer service systems” and the second is to connect these systems with local networks of citizen crisis responders.

How do large private sector companies deal with the social media challenge? Well, some obviously do better than others. (Incidentally, this question was a recurring topic of conversation at the Same Wavelength conference in London where I spoke after Geneva). This explains why I recommended that my ICRC colleagues consider various social media customer service models used in the private sector and identify examples of positive deviance.

The latest innovation in the customer service space was just launched at TechCrunch Disrupt this week. TalkTo “allows consumers to send text messages to any business and get quick responses to questions, feedback, and more.” As TechCrunch writes, “no one wants to wait on the phone, and email can be slow as well. SMS Messaging is a natural form of communication these days and the most efficient for simple questions. It makes sense to bring this communication to businesses.” If successful, I wonder whether TalkTo will add Twitter and Facebook to their service as other communication media.

Some companies leverage crowdsourcing, like Best Buy’s TwelpForce. Over time, Best Buy “found that with some good foundational guideposts and training tools, the crowd began to self-organize and govern itself.  Leaders in the space popped up as coaches, or mentors – and pretty soon they had a really good support network in place.”

On the humanitarian side, the American Red Cross has begun to leverage their trained volunteers to manage responses to the organization’s official Facebook page, for example. With some good foundational guideposts and training tools, they should be able to scale this solution. In some ways, one could say that humanitarian organizations are increasingly required to play the role of “telephone” operator. So I’d be very interested in getting feedback from iRevolution readers on alternative, social media approaches to customer service in the private sector. If you know of any innovative ones, please feel free to share in the comments section below.

The second strategy that humanitarian organizations need to consider is linking this new customer service system to networks of citizen crisis responders. An “operator” on the ARC Facebook page, for example, would triage the incoming posts by “pushing” them into different bins according to topic and urgency. Posts that don’t reflect a life-threatening situation but still require operational response could simply be forwarded to local citizen crisis responders. The rest can be re-routed to professional emergency responders. Geo-fenced alerts from crisis mapping platforms could also play an important role in this respect.

One should remember that the majority of crisis responses are “crowdsourced” by definition since the real first responders are always local communities. For example, “it is well known that in case of earthquakes, such as the one that happened in Mexico City, the assistance to the victims comes first of all from the other survivors [...]” (Gilbert 1998). In fact, estimates suggest that, “no more than 10 per cent of survival in emergencies can be contributed to external sources of relief aid” (Hillhorst 2004). So why not connect humanitarian customer service systems to local citizen crisis responders and thereby make the latter’s response more targeted and efficient rather than simply ad hoc? I’ve used the term “crowdfeeding” to describe this idea in previous blog posts like this one and this one. We basically need a Match.com for citizen based crisis response in which both problems and solutions are crowdsourced.

So where are these “new” citizen crisis responders to come from? How about leveraging existing networks like Community Emergency Response Teams (CERTs), the UN Volunteer system (UNVs), Red Cross volunteer networks and platforms like Red Cross Volunteer Match? Why not make use of existing training materials like FEMA’s online courses? Universities could also promote the idea of student crisis responders and offer credit for relevant courses.

Update: New app helps Queensland coordinate volunteers.

13 responses to “How to Crowdsource Crisis Response

  1. Hey Patrick! Many thanks for popping over to Geneva and #RCcom. It was a pleasure meeting you and we are all very excited about the work you and Ushahidi do in the field of #crisismapping. As discussed, it would be great if we can find some time when you’re back (Crisis Mappers Conference in November) to follow up on “Health Care in Danger”. So long until then and best of luck with your projects. Cheers /Isaac

  2. Hi Patrick, thought provoking post as usual!
    In relation to your question on how to reach people.
    One idea that I had was geared around Facebook because of its high degree of popn. reach, particularly in the US.
    (Whether or not they would do it is another question.)
    The idea came from the Tuscaloosa Tornado and goes like this.

    Facebook has a record of where you live, (even where you sign in via IP addresses) and when you were last active.
    It could release a broadcast message to all its users within a defined region as determined by the disaster. If they have their facebook connected to their phone the message will come down onto their phone via app, email or sms
    The Question “are u ok”? can we pass your details onto emergency services?
    Response
    “I am ok, please pass my details to the local EOC (could be Sheriffs office)”,
    “I am ok, pass my details only to my facebook friends”
    “No, I need help”
    The responses are recorded and a list of these responses can be passed over to local emergency services.
    Using checkin function determine responses that are out of impact zone
    These could automatically go up on a map with a colour grade

    The idea being that you can rapidly see where the people are what their status is
    and therefore deploy teams based on highest benefit

    Or something along these lines :)

    The point is, Facebook has a large database that could be used for quickly determining user status within affected zones. As a chunk of the population this is a good percentage accounted for

    Google + could also be suitable,

  3. I am thinking that this should be very possible – of course defining our exact tasks clearly and then building the right crowd [with the right skills/understanding/training] would be the challenge.

    At first, I was thinking that each [humanitarian] organization might want to build its own crowd for their specific task, but then realized that perhaps this could be part of the role of the Humanitarian Standby Task Force [i.e. building a humanitarian crowd that could be used for such tasks] which is part of the Communities of Interest that UN-OCHA is hoping/helping to move forward. If we had a vetted crowd [or a sub-group of the crowd], then any humanitarian organization could provide relevant training to that group…..and then that group could be activated by the organization during emergencies as needed.

    Thoughts?

    Cheers,
    Andrej

  4. Pingback: The Best of iRevolution: Four Years of Blogging | iRevolution

  5. Pingback: Crowdsourcing a Crisis Map of the Beijing Floods: Volunteers vs Government | iRevolution

  6. Pingback: How People in Emergencies Use Communication to Survive | iRevolution

  7. Pingback: How to Create Resilience Through Big Data | iRevolution

  8. BARUME BISIMWA ZIBA Chef d’Equipe CROIX-ROUGE Uvira sud-kivu en RDCongo demande de materiaux de Secours d’ Urgence . SOS Contact mail barume2008@yahoo.fr tel +243 971603199 +243853195164 contact Facebook barume bisimwa ziba Merci

  9. Pingback: Humanitarianism in the Network Age: Groundbreaking Study | iRevolution

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s