Stephen Thompson, Head of Digital at the University of Sheffield looks at why organisations are adapting their processes to incorporate social media for crisis communications. Wasim Ahmed, PhD candidate, provides a practical overview of the social media tools that can be used to monitor and track crisis communication issues that may arise.
Crisis communication issues can take many forms. Organisational issues such as the 2014 Sony Playstation hack saw the company criticised for not managing its social media. After Sony had released information related to the hack, Twitter users tweeted over 78,000 tweets in 16 hours. Crisis communication also comes to the fore in situations where human life is involved, such as in July 2013 when a Boeing 777 crash landed at San Francisco International Airport. A citizen photographed the event and uploaded the image to Twitter within less than one minute of the incident taking place. In just 30 minutes there were over 44,000 tweets related to the accident which included photos and videos which were taken by survivors as they escaped the wreckage.
Over recent years social media and crisis communication has drawn huge interest, and this year alone there are already over 77 thousand entries in Google Scholar of the keywords social media and crisis communication. Readers of this post may be interested in the research of Carlos Castillo, and Patrick Meier. Additionally, as shown in Figure 1 below Google trend web queries for the term ‘social media crisis’ have also risen over the previous 9 years.
Figure 1 – Time series graph of ‘social media crisis’ over the previous 9 years (image courtesy of Google Trends).
The above cases highlight that social media, and especially Twitter, is often the first place that you will see a report of an incident. Whatever the size or type of impending crisis, someone will post about it. A job role such as Head of Digital at a university can involve potentially having to deal with both organisational issues as well as those potential crisis events which concern human life.
Typically, the first step is to begin assessing the validity of a stream of messages (often drawing on the automation of the tools listed in Table 1) and posting a holding message that shows concern, the action you are taking and finally providing reassurance. This is not always the easiest to achieve within a single tweet, nor entirely possible at the first stage of a crisis. At this stage, an account that may be posting about a crisis would be reviewed and other posts and social media channels would be checked for similar reports.
In the context of a University this means checking geographically ring-fenced accounts such as those deriving from Yik Yak and using tools such as Echosec, which is a location based social media platform. Evidence that an incident is occurring will also be passed onto relevant stakeholders and incident procedures will take place.
From then on, social media would ideally be used to proactively post significant updates to followers, reply to individual posts and provide real time insights to the incident team. However, it is worth noting that this data will only provide an indication of what is occurring. Therefore, when replying to users and posting updates, consideration may be given to how other members of the public, and the media may pick up the responses and use these as part of relaying the story themselves. It is important to be conscious that tweets or posts may be used in media stories and this might not be linked to a media statement that is provided.
Once the crisis subsides, a review and evaluation of event(s) would take place which would aide identifying possible areas for improvement in the management of the crisis, or the product or service offered. Within this evaluation of events, social media posts may be analysed to identify any key areas of where a crisis situation changed and how responses could have been more effective. The above steps take into account i) showing concern, ii) taking action, and iii) finally providing reassurance.
Those wishing to capture and analyse or merely monitor Twitter during crisis communication events will require almost real-time access to the data. There are several free open source tools, and commercial tools which can be used by academics. We have only included software below that we have used and which we have used in a crisis communication situation.
Teams responsible for social media in a crisis situation will be actively monitoring platforms such as Twitter, Yik Yak, Facebook, and Periscope. In regards to the management of social media data, tools such as Hootsuite and TweetDeck can become very handy. The tools listed below can be used by those working within the Digital Media sector across a variety of organisations, but also by researchers looking to gain a better understanding of viral stories.
Table 1 – Tools that can be used for handling social media crisis situations
|Tool||Operating System||Interesting feature(s)*||Free or Paid?|
|Echosec||Web Based||Allows you to search social media by location i.e., a particular geographical area such as a university.||Free and pro versions available|
|NodeXL||Windows||Allows the ability to visualize networks of social media data and produce reports with top influencers, keywords, and hashtags.||Academic and commercial licenses available|
|Visibrain||Web Based||Provides real time data of crisis situations on Twitter and can handle huge volumes of data.||Free 30 day trial available|
|Pulsar||Web Based||Provides ability to monitor large volumes of data and examine location and clusters||Demo available|
|DiscoverText||Web Based||Allows the ability to import data via Sifter, and analyse crisis situations historically.||Demo available|
|Follow the Hashtag||Web Based||Allows free (short term) access to monitor crisis situations.||Free and paid versions available|
|Twitonomy||Web Based||Provides the ability to monitor crisis situations via examining mentions of a Twitter account.||Free and paid versions available|
|Chorus||Windows||Allows the ability to download and analyse Twitter data via the Search API.||Free|
|Mozdeh||Windows||Allows the ability to download and analyse Twitter data via the Search API.||Free|
|TAGS||Web Based||Allows the ability to use Google Spreadsheet to quickly set up data retrieval via Twitters Search API.||Free|
|COSMOS||Windows and MAC OS||Runs on MAC OS, and uses the Streaming API. Allows ability to create geo-maps of tweets.||Free|
|Hootsuite||Web Based||Allows the ability to monitor and respond to posts across multiple social media platforms, review post response times||Free and paid versions available|
|Tweetdeck||Web based||Provides the ability to monitor and respond to tweets||Free|
*As defined by our use cases
If you are Interested in finding out more about relevant crisis communication issues, we recommend taking a look at a recent @NodeXL network report created by Marc Smith which maps the keyword ‘crisis communication’ which displays the most frequently used URLs and topic discussions.
Featured image credit: irfanahmad1989 (pixabay public domain)
Note: This article gives the views of the author, and not the position of the LSE Impact blog, nor of the London School of Economics. Please review our Comments Policy if you have any concerns on posting a comment below.
Wasim Ahmed is a PhD candidate at the Information School, at the University of Sheffield.. Wasim has a very successful research blog which includes posts about key trends and issues within social media, but also covers more practical posts on using tools to capture and analyse social media data. Wasim is a keen Twitter user (@was3210), and will be happy to answer any questions you may have.
Stephen Thompson is Head of Digital at the University of Sheffield, and his role involves overseeing central University social media accounts, as well as website and video content. This includes preparing for and managing the communications during incident and crisis situations. Steve tweets @ratamahattass and is always open to questions and new ideas.