A theoretical and empirical study of social media as a crisis communication channel in the aftermath of the BP Gulf of Mexico oil spill

Kylie Ann Dowthwaite

Student thesis: Master thesis


On the 20th of April 2010 the largest oil spill in the history of the US occurred and BP was the main responsible party. This lead to a massive crisis communication effort, where multiple organisations were involved and social media was utilized on a scale never seen before in BP. Within classical crisis communication the Situational Crisis Communication Theory (SCCT), by Coombs, states that an appropriate response strategy has to be chosen depending on the crisis responsibility of the organisation. In the field of crisis communication for social media the theoretical framework is sporadic and very operational with Gonzalez-Herrero & Smith and Conway et al. being the main scholars identified. As the theoretical framework for social media as a crisis communication tool is new and sporadic the researcher dared to develop a holistic model, the KAD model, addressing this subject. Such a holistic model is believed to make it possible for an organisation to diminish the reputational impact of a crisis. The model ties current research together into a framework of mesco (organisational), micro (message) and macro (societal) dimensions. To obtain empirical data the researcher collected quantitative as well as qualitative data. A survey was sent out to 171 potential respondents predominantly in North America questioning the respondents on BP’s use of social media and their attitudes towards BP in relation to the oil spill. The survey response rate was 36%. Furthermore five key informant interviews were conducted with former high ranking stakeholders in BP’s communications, federal & public organisations and the seafood industry. The findings of the theoretical and empirical research showed that at the onset of the crisis BP did an extremely poor job with their social media efforts, although once they ramped up their social media channels they performed well, on a practical level. However it was found that more engagement (two-way communication) was needed as well as greater information tailored towards the needs of social media community, in order for communication to render effective. Furthermore it was clearly shown in the data, that the communications were hindered by the stigma BP obtained due to not projecting the same communication strategy throughout the response, as depicted by SCCT. This subsequently eroded the effectiveness of the overall response including online efforts. It is the view of the researcher that if BP had applied a more holistic view of social media use within crisis communications, such as the KAD model, they would have been able to better manage their reputation through social media channels during the crisis.

EducationsMSc in Brand and Communications Management, (Graduate Programme) Final Thesis
Publication date2011
Number of pages164