Uses of Social Media to Inform Operational Response and Recovery During an Airport Emergency (2017)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

35 CHAPTER FOUR FINDINGS, CONCLUSIONS, AND FURTHER RESEARCH PROFILES OF SOCIAL MEDIA FOR EMERGENCY MANAGEMENT PROGRAMS The key characteristics of the six case examples are illustrated in Table 7. Examination of the patterns in the table suggests that social media for emergency management (SMEM) programs that are believed by their owners to be successful have highly similar features. The six organizations in the case examples have nearly identical profiles. The few isolated differences result from the stage of maturity of a program or from restrictions placed on the program. An example of the first is Min- neapolis–Saint Paul International Airport’s (MSP’s) relatively limited range of uses of information from social media (SM), which apparently results from the newness of that airport’s SMEM program. Nevertheless, the #BlackLivesMatter experience recounted in the case example suggests that the airport has used SMEM inputs for decision making and resource allocation although MSP did not note such uses on the survey. As an example of continually deepening the use of SM, MSP used it almost exclusively as the focal point of its 2016 emergency exercise of integrating its SM into the joint information center and emergency operations center (EOC) during an emergency. An example of the second is the decision by the New York City Office of Emergency Management (NYC OEM) not to seek conversations in its SMEM channels. This preempts crowdsourcing. The choice results from a desire to avoid the creation of new liability issues and to avoid undercutting existing pathways such as 911 for requesting aid and reporting emergencies. There is a strong preference for buying or subscribing to social listening software and running it in house. The six success- ful programs appear to have the same persons running the software and doing human evaluation of the outputs. “Running” the software means setting the search or alert criteria in the software and deciding which information to pass to the EOC and decision makers. The staff sizes in the case examples are consistently small, except for Southwest Airlines (SWA). However, the staff esti- mates for the other five programs are for SMEM whereas the SWA estimate is for the airline’s total SM program including SMEM, branding, brand protection, and customer service. The size of SWA’s program approximates that of two other large corporations for which partial information was obtained but not approved for use in this study. The SMEM staffs for the other five programs are small because their total staffs are so much smaller than SWA’s 53,000 employees, and their communica- tions staffs are cross-trained to move into SMEM roles during events. Approximately 0.08% of SWA’s total employees are in the SM unit; as are 0.46% of MSP’s total employees (three of 650). Most of the programs use on-the-job training, which is reasonable given the relatively small staff sizes and the unlikelihood of cohorts of new hires to take a formal course. However, the American Red Cross trains its volunteers and the University of North Dakota (UND) has a formal training class (internally developed) for police officers and communications staffers who perform their SMEM functions. The UND lesson plan is reproduced as Appendix C. The recommendations made in FEMA course IS-042 are reproduced as part of Appendix F to this report. The costs of the programs are not shown in Table 7, as directly comparable data were not collected from every cases. The costs seem to range from near zero for the American Red Cross (software and social listening services are donated and Digital Disaster Operations Centers use volunteer labor) to approximately $650,000 per year for NYC OEM [$150,000 for the depart- ment’s share of the city’s subscription to the social listening software plus 10 full-time equivalents (FTE)]. For comparison, Edmonton International Airport’s (EIA’s) annual cost for SMEM is approximately $66,000 ($6,000 for software and social listening services subscription plus $60,000 for 0.50–0.75 FTE for SMEM). As a rule of thumb, a fully realized SMEM pro- gram appears likely to cost about 1% of an organization’s total annual operations budget. An airport can avail itself of one of the less expensive software programs or obtain SMEM services from its city, county, or a regional organization.

36 TABLE 7 SELF-REPORTED CHARACTERISTICS OF SMEM PROGRAMS AT CASE EXAMPLE ORGANIZATIONS EIA MSP NYC SWA UND YVR SMEM used for notifications SMEM used for warnings SMEM used for situational awareness SMEM used for resource allocation SMEM used for decision making SMEM used for investigations SMEM uses crowdsourcing (outgoing request for information) Has written policy governing SMEM program, its use, and its control Under development Clearly stated goals for SMEM program Clearly assigned responsibility for control and mainte- nance of SMEM SMEM integrated with organization’s overall SM program Partly SMEM part of organization’s comprehensive crisis com- munications plan Tight control over outgoing messages Use of pre-scripted messages Partly Organization seeks engagement (“conversations”) through SMEM Uses social listening/social monitoring Uses social listening based on key words, key phrases, hashtags Uses geofencing ? Uses commercially available software Runs social listening for self (in house) Subscribes to data-bundling service Monitors 24/7/365 Stands up only for disasters Dedicated staffing Staff size (FTE) (employees) for SM (all SM, not just SMEM) 0.75 3 10 31-45 3-5 6 Formal training program for SMEM staff On-the-job training for SMEM staff Uses trained volunteers To surge Uses automated analysis Uses human analysis Uses hybrid analysis SMEM inputs used in EOC Communications group (SM/SMEM) representative in EOC when EOC stands up Automated direct display of SMEM inputs in EOC On demand On demand On demand On demand On demand On demand Human-selected display in EOC On demand On demand On demand On demand On demand On demand Automated decision making in EOC or elsewhere based on SMEM inputs Organization happy with its SMEM Source: Smith and Kenville data. Note: grey = yes; light grey = qualified yes; medium grey/? = unstated but probably yes; black = no.

37 CHALLENGES Four challenges came up in the case examples and in the literature: 1. Gaining senior management support for allocating resources to SMEM; 2. Gaining support among emergency managers and other first responders for use of SMEM, particularly the application of information from SMEM to resource allocation and decision making (a cultural shift for most emergency manage- ment personnel to rely on SM inputs); 3. Validation and verification of incoming information from social listening; and 4. Choosing the right software packages for social listening and analysis and deciding which social media streams to listen to. With regard to gaining senior management support, the consensus of the case examples is that support from senior man- agement is essential to adequately resourcing SMEM and that nothing is as convincing as success. Since most of them had already had success using SM for branding, brand protection, and customer service, building SMEM onto the general SM program pre-sells SMEM with senior management. Once SMEM is activated, it impresses senior management by producing results at a very low cost. The case examples point out that SMEM applications of their SM platforms increase the power and reach of the organizations’ overall presence in the media space and with the public. Once begun, SMEM is self-reinforcing. Winning over emergency managers and other first responders requires patience and modesty on the part of SMEM practi- tioners. Information from SMEM is just another information stream. It needs to be moderated to avoid overloading decision makers and staff in the EOC. The case examples have given this moderating function to communications professionals who typically have a presence (in person or with an on-demand display) in the EOC. Effective programs use manual evaluation or hybrid evaluation of the SMEM inputs. Social media have power because they are open for anyone to upload posts, images, or videos. This is also the biggest shortcoming of social media as an SMEM tool. The outputs of social listening software, at least at the present state of develop- ment, need human efforts to verify and validate. Such human efforts appear to be facilitated by visual or graphic presentation of SMEM outputs rather than text or written outputs. There is a wide range of commercially available software available for social listening. Airports can work with vendors to find programs that fit their needs and budgets. The choice of what is most appropriate goes back to the airport’s goal in apply- ing social media to emergency management. CONCLUSIONS Assuming these foundational principles, • The influence of social media is still increasing, which means the potential usefulness of SMEM is also increasing; and • Applying information gained through social listening and analysis of social media for emergency management can benefit airports, but the staffing and other cost concerns may limit the applicability to small airports (the smallest orga- nization in this study is the equivalent of a medium hub airport), the following conclusions may be drawn from the case examples and the literature review: 1. Having a clear vision of what the airport wishes to achieve through SMEM is the essential first step; everything else should follow. 2. An airport with a social media presence and in-house experts can be ready and valuable to support emergency respond- ers in any event.

38 3. Growing social media and SMEM programs internally seems most beneficial. This process works best when an airport does SMEM for itself rather than outsourcing part of all of its SMEM program. The employees working for that entity can give it the time, knowledge, and care it needs to grow. – Most organizations utilize on-the-job training, as the SM landscape evolves quickly. – Rotating workers avoids SM burnout and empowers the employees so there is no appearance of being scripted or nonresponsive—timely is the name of the game. This applies when the program seeks to be interactive (i.e., have conversations) and even when the program uses pre-scripted responses in some situations. 4. A designated SMEM champion can show the value of SM programs to the C-level in the organization. This requires determining what metrics will be utilized to show the value of the SM presence and win more resources as needed. – Participation from other departments across the organization is key; this enables the SM presence to be valued by all and allows for resources to be shared. – Engaged employees (“evangelism” as the Los Angeles Fire Department and Southwest Airlines term it) will enhance overall employee morale. Seeing their organization’s efforts in responding to or leading an issue will be a source of pride. 5. Organizing and staffing one’s SMEM on the basis of desired communication outcomes is far superior to starting from a technological or outside cultural standpoint. Those with experience in customer service and marketing can use SM and SMEM analytical tools to provide decision-quality information and actionable intelligence. – A thoughtful SM presence will build creditability for the organization. – Superior emergency management as enabled by SMEM will be viewed as part of an airport’s overall superior reputa- tion and customer service. – An effective SM and SMEM platform can serve the broader community (e.g., the EIA case example). 6. Embedding the SMEM program into the airport’s overall crisis communications strategic plan and SM program is productive. 7. The public information officer and emergency managers need to develop a relationship that promotes information shar- ing and joint planning, training, drilling, and exercises. FURTHER RESEARCH The findings of this synthesis suggest that further research is needed on the following topics: • A guidebook for the creation and maintenance of social media for emergency management systems for airports of all types and sizes; • Model pre-scripted announcements, posts, and so forth for airports to use in SMEM notifications, warnings, and outgo- ing information; • Evaluation of current validation and verification systems for SM-derived information including automated systems; and • Update of ACRP Report 94 (Integrating Web-Based Emergency Management Collaboration Software into Airport Operations—A Primer) and ACRP current research project 16-04-03 (Guidebook for Airport Emergency Operations Centers) to account for display and use of SMEM inputs in emergency operations centers.