This exploratory study aims to investigate social media users’ reactions via user-generated content (UGC) to crisis-affected supply chains, through the lens of Situational Crisis Communication Theory, to: (1) ascertain whether the Social Media Users' Crisis Response (SMUCR) Framework is applicable to a prolonged, multi-brand, global crisis and (2) whether there is any evidence of online brand advocacy (OBA) and online brand detraction (OBD).
Natural language processing (NLP) with Leximancer, enabled thematic and sentiment analyses of 295,024 X (Twitter) posts extracted over a three-year period.
This study found that there were nine stages in social media users’ response to a long, global supply chain crisis. It also found sentiment coupling as positivity and negativity were not mutually exclusive and co-appeared in the UGC throughout the 3 years. However, not all positive sentiment demonstrated OBA dimensions and not all negative sentiment mirrored OBD at various stages of the crisis.
This study enhances the SMUCR Framework by incorporating the evolving role of social media users in shaping brand narratives during crises through OBA and OBD. It highlights the fluctuating nature of public sentiment, showing how consumer voices influence brand perception online over time. This study updates the SMUCR Framework from four to nine social media users’ crisis response stages. It sheds new light on the role which social media users play in crisis evolution and management online.
The updated SMUCR Framework will enable industry practitioners to better anticipate, manage and respond to an elongated, global crisis which evidences itself via social media UGC. A fresh perspective is provided on crisis management, stressing the need to monitor and adapt to changing social media discourse to sustain brand resilience to a crisis.
This paper extends the original SMUCR Framework beyond a one-brand, short-term crisis scenario through a multi-brand, longitudinal, global crisis lens and evolves the Framework from four to nine stages.
Introduction
Through their user-generated content (UGC), social media users are powerful voices of what is perceived by their connections as “truth” (Luo et al., 2022). At a time of crisis, consumers turn to their social networks for support, information and social interaction. This was evidenced during the recent COVID-19 pandemic, which saw consumers turn to social media to seek advice and support in coping with supply chain issues evidenced by empty supermarket shelves (Li et al., 2023; Wilk et al., 2022a, b, 2023).
According to Coombs (2015), a crisis is “the perception of a predictable event that threatens important expectancies of stakeholders related to health, safety, environmental, and economic issues, and can seriously impact an organisation’s performance and generate negative outcomes” (p. 3). Notably, during the COVID-19 crisis, supply-chain management has been severely implicated, with van Hoek (2020) pointing out that more than 86% of supply chains were affected. The negative consumer responses to the global shortages in stores contributed to panic-buying during the crisis (Wilk et al., 2022a, b, 2023). Researchers have noted that, from a supply chain management perspective, the highly influential public opinion of social media users can have an impact on the management of a crisis. Consumer-driven narratives provide insights into fluctuations in marketplace conditions such as recessions which affect consumers expectations, preferences and values (Drewniak and Karaszewski, 2016) and are valuable in informing the efficiency of supply chain management planning and operations, as well as strategising (Palalic et al., 2021). Recent studies have reported that social media users were, for the most part of the COVID-19 pandemic, the only source of information which consumers could rely on during the supply shortages and that despite all the negativity about supplies, there was also positivity evidenced in online narratives (Wilk et al., 2022a, b, 2023).
Wilk et al. (2023) and Prins and Wellman (2021) provide insights into positivity amongst negativity but encourage others to replicate their findings by exploring social media narratives in the crisis context over a longitudinal, multi-year period. Additionally, Wilk et al. (2023) note that UGC may contain consumers’ thoughts, attitudes and emotions which contribute towards brand sentiment and need to be effectively managed to create a positive brand image during a crisis (Kall et al., 2006; Drewniak and Karaszewski, 2016). The study of Drewniak and Karaszewski (2016) found that the market orientation and flexibility of brands during a crisis have not yet been effectively investigated. Notably, these strategic marketing concepts can be linked to the Social Media Users’ Crisis Response Framework (SMUCR)’s stages (Morgan and Wilk, 2021) as a brand’s external stakeholders active on social media during a crisis can help strengthen a brand’s overall competitive positioning in the market. It has been reported that positive UGC can have the effect of attracting consumers to a brand, while negative UGC can result in brand avoidance (Azer and Alexander, 2020). Morgan and Wilk (2021) found that social media users participate in crisis management via their UGC to their online networks and during their exchanges, they spread positive and negative sentiment to others. Specifically, these researchers have found that there are four stages which form, what they term, the SMUCR, which social media users seemed to follow in a 12-month sport-related crisis. However, their Framework has not as yet been validated in any other context nor in an elongated global crisis, presenting thus an opportunity for this study to address this gap.
Furthermore, consumers’ social media exchanges, give rise to positivity and negativity about brands, and this is particularly important to understand during a crisis which has the potential to affect the longevity and ultimate survival of a business. Consumers can speak up for brands, share detailed brand knowledge and defend and support the brand in a time of need giving rise to online brand advocacy (OBA) (Wilk et al., 2019, 2020, 2021). Conversely, customers can spread negative brand experiences to their social networks via sharing extremely negative opinions and degrading a brand in an effort to detract others from the brand in their decision-making. Such actions are known as online brand detraction (OBD) (Wilk et al., 2022a, b).
Wilk et al. (2018, 2020) introduce OBA and highlight that OBA is an under-researched construct, requiring further contextualisation. This is because both OBA and OBD are important expressions found in UGC as both are perceived as the most authentic and trustworthy voice by customers of a brand online and thus may have the power to alter the way a brand crisis evolves within online networks (Wilk et al., 2022a, b, 2023). Despite scant evidence in extant research papers on this phenomenon, we hypothesise that both, OBA and OBD, are present in UGC during a global, multi-brand crisis.
Customer-centrism in supply-chain structures can occur by businesses understanding their customers’ purchase behaviours and patterns, such as noticing perceived significant differences between products depending on their level of product involvement (Palalic et al., 2021) and platform-mediated communications (Zinnbauer and Eberl, 2005). In additive manufacturing the organisation of the manufacturer’s supply chain is becoming less hierarchical in structure and more decentralised, which is vastly shifting the role of the consumer as an active contributor to the value-creation process (Bogers et al., 2016). Furthermore, supply chain managers and businesses are required to respond swiftly to brand-related crises. Researchers have urged further, longitudinal investigation in this context, as at present there is limited research into the influence of social media UGC on consumers’ issues or opinions on products and service information and experiences relating to supply chain management in real-time (Palalic et al., 2021). Our research seeks to provide these longitudinal insights.
Therefore, through the lens of Situational Crisis Communication Theory, this study seeks to identify the applicability of the SMUCR Framework developed by Morgan and Wilk (2021) beyond a one-brand, short-term crisis scenario through a multi-brand, longitudinal, global crisis lens, to see if the Framework needs to be extended to a difference scenario, such as the COVID-19 crisis. This is important as an SMUCR framework should encompass a wide range of user behaviours and responses and nuances applicable to a raft of crises, which perhaps a four-step model may have missed. Since different crises trigger varied responses, an expanded framework might need to be designed to be more adaptable across different scenarios, making it a valuable tool for crisis management. Such expansion would, therefore, incorporate the latest research on social media behaviour during a crisis. As our understanding of social media dynamics evolves, the extant SMUCR framework must be tested and possibly refined to reflect new insights (Morgan and Wilk, 2021). In such refinement, this study aims to better understand to what extent OBA and OBD evidence themselves in social media users’ narratives about a supply chain crisis, and what role, if any, OBA and OBD play in crisis management during the pandemic. This study seeks to extend the studies of Wilk et al. (2022a, b, 2023) into the sentiment of social media responses in the supply-chain context, by applying a longitudinal UGC analysis over a three-year period.
Thus, the three research questions which this study seeks to answer, are:
Do the social media UGC about supply chain issues follow the stages outlined in the Morgan and Wilk (2021) SMUCR Framework?
Are there any characteristics of OBA present in social media UGC about the supply chain crisis?
Are there any characteristics of OBD present in social media UGC about the supply chain crisis?
Literature review
Situational crisis communication theory (SCCT)
Coombs (2022, 2007) posits that organisations should consider consumer emotions that are vested in a brand during a crisis, as the powerful influence of consumer emotions can be hard for organisations to recover from. Situational Crisis Communication Theory (SCCT) postulates that the type of crisis response strategy should be selected based on the type of crisis, the severity of consequences, the performance history of the company, and the stakeholder’s attribution of the crisis responsibility (Coombs, 2004, 2015). Morgan and Wilk (2021) note that when an unexpected or negative situation arises, stakeholders can act to attribute blame in their search for the cause of the event. When assessing responsibility, the stakeholders assess whether or not the organisation could control the situational factors at hand (Coombs, 2007). In unprecedented situations of environmental influences (i.e. natural disasters), the organisation is often viewed as a victim of the event, with stakeholders directing very weak attributions of crisis responsibility (Coombs, 2007). Social media users can have an impact on the management of a crisis, and this is because consumers engage and conform with those who emit similar behaviours and characteristics as themselves, echoing the opinions of each other. Conversely, Morgan and Wilk (2021) point out that research to date focuses on the brand as the instigator of crisis response, rather than the consumer who uses social media to fuel brand-related narratives during a crisis, highlighting that businesses have neglected to incorporate the social media user in their crisis management strategies. There exists scant research on the social media users as instigators of crisis response on behalf of a brand, presenting a gap for this study to address.
Social media users’ crisis response (SMUCR) framework
The SMUCR Framework presents a lens through which corporate strategies can incorporate social media users in crisis management. Searches for citations for the SMUCR Framework on Google Scholar, Scopus, Emerald, Web of Science, Science Direct and Taylor and Francis, revealed that despite 28 papers citing this research, no studies to date have validated this work by Morgan and Wilk (2021). Morgan and Wilk (2021) state that the applicability of the SMUCR Framework should be examined in different settings and that research exploring online crisis response would be a valuable contribution to crisis communication. Stage 1 of the SMUCR is “Framing and Identity Development”. In the immediate period of the crisis, identity was given to the crisis situation by assigning a descriptive name or label to it – forming a collective distinctiveness among social media users, which created an identity and an underlying sense of commonality within the social media community. This stage included developing an online tribe, via OBA – the proliferation of positivity, information sharing, cohesiveness and defence. Villanising was also a trait of this first stage as social media users formed negative attitudes towards those individuals whom they identified as culprits (Morgan and Wilk, 2021). Stage 2 is the “Sense-making” stage where individuals retrospectively consider what happened and who was responsible for the crisis (Morgan and Wilk, 2021, p. 10). Stage 3 of the SMUCR is called “Accountability”, whereby social media users’ lack of protection of the brand during a crisis was damaging to the organisation, as users perceived the crisis as intentional which spurred negative emotions, particularly when third-parties were involved (Kim and Cameron, 2011; Morgan and Wilk, 2021, p. 11). Notably, Stage 4 included “Vigilant Rebuilding” as the crisis dissipated and social media users moved on to acknowledge what had happened and the fall out, and cautiously moving forward with optimism for the future. The SMUCR Framework by Morgan and Wilk (2021) is yet to be validated via application in contexts other than sport and with longitudinal data for an elongated global crisis, which this study seeks to address. We thus posit that the nature of the crisis, such as its longevity, location and number of brands involved, may result in a more elaborate crisis response by social media users. Therefore, our hypothesis is that in an elongated, complex, global and multi-brand crisis, there are likely to be more than the four stages outlined by Morgan and Wilk in their 2021 study, and we set out to present insights in this regard as follows:
In an elongated, global and multi-brand crisis, the SMUCR Framework will involve more than the four stages outlined by Morgan and Wilk (2021).
Online brand advocacy (OBA)
The holy grail for businesses is organic, earned UGC in which consumers advocate for brands to their online networks out of their own free will, unprompted and unincentivised (Wilk et al., 2020). OBA often evidences itself within online communities of active, loyal and increasingly connected consumers, who operate across multiple digital touch points (Tuten and Solomon, 2018). Wilk et al. (2020) explain that prosumers simultaneously co-construct meaning online by producing, consuming and creating brand-related content and their UGC is highly visible to their networks (Wilk et al., 2023). Vocal consumers resonate with likeminded individuals and are informed by each other’s brand-related knowledge which is shared on digital platforms, such as social publishing and e-commerce online shopping sites (Smith et al., 2012), social networking sites (e.g. Facebook or X), online opinion platforms (e.g. tripadvisor.com) and discussion forums in online communities (e.g. epicski.com) (Wilk et al., 2018, 2019). Wilk et al. (2019) characterise such UGC as OBA arising from a customer’s strong, powerful, deliberate and non-incentivised online representation of a brand and its best interests. Online brand advocates are powerful representatives of the brand who voice their views to stand up for the brand by supporting or defending against any negative claims about a brand during a crisis to other consumers, which can restore a positive effect for the brand in a time of need (Wilk et al., 2019, 2021).
Wilk et al. (2021) suggest that there are four dimensions of OBA, namely: brand knowledge, brand defence, brand positivity and virtual visual cues. In an economic crisis situation, brand positivity can be fostered by creating an appealing brand image, specifically relating to physical and intangible attributes within its product category, at the corporate or individual level of consumers’ needs, thoughts, opinions and desires to a prospective buyer to create brand value and competitive differentiation (Kall et al., 2006; Drewniak and Karaszewski, 2016). Despite brand advocacy being primarily driven by the customer (Aksoy et al., 2023; Wilk et al., 2021; Wong, 2023; Wong and Hung, 2023), it also includes other stakeholders, such as employees (Badrinarayanan and Sierra, 2018) and suppliers (Leach et al., 2021).
In line with, Coombs (2007) research, this study proposes that if stakeholders within the organisational context can be advocates, supporting customer-centred relations (Leach et al., 2021), then they may also form an understanding of how to best support the necessity and positivity (Drewniak and Karaszewki, 2015) of brands from the viewpoint of consumers, who engaged in panic-buying in the COVID-19 context. We therefore posit that OBA is likely to be present in social media users’ responses to a crisis, as social media consumers are likely to share their brand experiences, knowledge and brand defence with others, however, social media users may not be able to share OBA throughout all stages within the SMUCR Framework, as the brand knowledge and experience acquisition may not be immediate (i.e. it takes time for COVID vaccinations to be developed, supplied, administered and experiences). Our hypothesis is as follows:
OBA is present in social media users’ responses to a crisis but not in all stages of the SMUCR Framework.
Online brand detraction (OBD)
OBD is founded on a broken customer-brand relationship and typifies a strong negative online brand behaviour of detracting others from a brand (Wilk et al., 2022a, b). It is a customer’s negative emotional state toward a brand that has the potential to damage a brand’s reputation (Kucuk, 2016). Alike brand hate, customers silently and passively distance themselves from the brand and use aggressive forms of conduct to demonstrate their anger and public outcry towards the despised brand. Such brand detractors are known to discourage others from a brand via proliferating negativity on social networks (Wilk et al., 2022a, b). Brand detractors epitomise consumer decisions in cancel culture and in negative consumer behaviour, such as brand dislike and brand avoidance. An exploratory study in the context of an online opinion platform, found four key dimensions of OBD: dissuading others from the brand, negatively-framed knowledge sharing about the brand and comparison with others, attacking the brand and negatively themed virtual visual cues (such as links and hashtags) (Wilk et al., 2022a, b). OBD is said to evidence itself as having the polar opposite dimensions to those found in OBA (Wilk et al., 2022a, b). It is important to manage evidence of brand hate, online detractors and consumers’ negative attitudes to brands at a firm level during a crisis, as these can have a detrimental effect on the brands’ position in the marketplace and brand loyalty (e.g. Drewniak and Karaszewski, 2016). Prior studies in this context posit that to attain a favourable market orientation during a crisis, firms should focus on strengthening their market position by delivering what the brand promises with a strong brand identity that aligns with consumer expectations. Balancing the emphasis on a quality orientation with transparency has been shown to result in high brand credibility (Drewniak and Karaszewski, 2016). The extreme form of customer-driven negativity online, namely OBD, has not yet been investigated in the context of a crisis or in supply chain management.
We posit that OBD is likely to be present in social media users’ responses to a crisis, as social media consumers are likely to share their negative brand experiences and brand attacks, however, social media users may not be able to share OBD throughout all stages within the SMUCR Framework, as the brand knowledge and experience acquisition may not be immediate (i.e. it takes time for COVID vaccinations to be developed, supplied, administered and experiences). Therefore, our hypothesis is as follows:
OBD is present in social media users’ responses to a crisis but not in all stages of the SMUCR Framework.
Key literature underpinning this study is presented in Table 1. Consequently, this study set out to better understand how OBA and OBD evidence in social media users’ crisis responses in the context of the COVID-19-affected supply chains. Specifically, this study sought to apply the SMUCR Framework to identify the extent to which it can be applied in a long-term, global crisis situation.
Summary of key literature
| Construct/framework | Definition | Empirical/theoretical/conceptual | Context | Key findings | References |
|---|---|---|---|---|---|
| Online brand advocacy (OBA) | “OBA is a strong, influential, purposeful and non-incentivised, online representation of a brand and a brand’s best interest, by a brand-experienced customer (either past or current), which includes ‘standing up for’ and speaking on behalf of the brand. OBA is given as UGC that reflects specific and in-depth brand knowledge, defence, positivity and virtual positive expression in support of the brand. OBA is a multidimensional construct that may be simultaneously positively and negatively valenced” (Wilk et al., 2020, p. 417) | Empirical: multi-methods, scale development | Online | 16-item, four-dimensional OBA scale was developed. OBA was found to be a multidimensional construct with four dimensions (i.e. brand defense, brand information sharing, brand positivity and virtual positive expression). Further, this study showed that OBA is conceptually different to consumer-brand engagement and eWOM | Wilk, V., Soutar, G., and Harrigan, P. (2020). Online Brand Advocacy (OBA): The development of a multiple item scale, Journal of Product and Brand Management, 29(4), 415–429 |
| Proactive consumers, prosumers, simultaneously co-construct meaning online by producing, consuming and co-creating brand-related content which is highly visible to their networks and which contains OBA (Wilk et al., 2023) | Empirical, qualitative | Online | Positivity via OBA was present in social media users’ narratives about supply chain shortages during COVID-19 | Wilk, V., Roni, S.M. and Jie, F. (2023), “Social media users’ perspectives on the supply chain: positivity amongst negativity during COVID-19 panic buying in Australia”, International Journal of Agile Systems and Management, Vol. 16 No. 2, pp. 262–287, doi: 10.1504/ijasm | |
| Online brand advocates are powerful representatives of the brand who voice their views to stand up for the brand by supporting or defending against any negative claims about a brand during a crisis to other consumers, which can restore a positive effect for the brand in a time of need (Wilk et al., 2019). A prosumer typically gives OBA as they simultaneously produce and consume online content at and by doing so can advocate for brands online (Wilk et al., 2019, p. 96) | Empirical, qualitative | Online | Consumers give OBA by comparing and contrasting competing brands against the brand they favour | Wilk, V., Soutar, G. N., and Harrigan, P. (2019). Tackling social media data analysis: Comparing and contrasting QSR NVivo and Leximancer. Qualitative Market Research: An International Journal, 22(2), 94–113 | |
| Brand management | “Brand management is creating brand awareness and image, by creating an identity that reflects the ideal image of the brand” (Drewniak and Karaszewki, 2015, p. 744) | Empirical, quantitative | Offline | Brand image has a strong impact on brand-loyal behaviour such as OBA. Brand’s leadership position is maintained for some companies that have experienced brand popularity via OBA over time | Drewniak, R., and Karaszewski, R. (2016). Brand management in a situation of an economic crisis: methods of strengthening the brand value in the scope of emerging markets. Asia Pacific Journal of Marketing and Logistics, 28(4), 743–758 |
| Online brand detraction (OBD) | OBD negativity occurs when users share their “undesirable brand experiences, insights, and opinions to their online networks through user-generated content others online” (Wilk et al., 2022a, b, p. 1) Four key dimensions of OBD include: brand negativity, brand attacks, negative brand knowledge and experience sharing and virtual visual cues (Wilk et al., 2022a, b) | Empirical, qualitative | Online opinion platforms and negative word of mouth on social media (Wilk et al., 2022a, b) | Both OBD and OBA are required for authentic social exchanges on Online Opinion Platforms (OOPs) Brand detractors are known to discourage others from a brand via proliferating negativity on social networks (Wilk et al., 2022a, b) | Wilk, V., Lambert, C., and Meek, S. (2022). Online brand detraction in an online opinion platform. International Journal of Hospitality Management, 107, 103345 |
| Situational Crisis Communication Theory (SCCT) | SCCT explains how the type of crisis response strategy should be selected based upon the type of crisis, the severity of consequences, the performance history of the company and whom the stakeholder’s attribute the source of blame towards for the unexpected crisis (Coombs, 2004) Situational Crisis Communication Theory argues that persons’ knowledge of a “past crisis” helps crisis managers “adjust their communication to account for possible past crises which relevant publics ae aware” (Coombs, 2004, p. 227) | Empirical, quantitative | Offline | A link exists between crisis history and perceptions of crisis responsibility; whereas, there is a strong link between crisis history and organisational reputation in victim or accident (moderate threat) clusters | Coombs, W.T. (2004), “Impact of Past Crises on Current Crisis Communication”, Journal of Business Communication, Vol. 41 No. 3, pp. 265–289, doi: 10.1177/0021943604265607 |
| Situational Crisis Communication Theory provides an evidenced-based framework for understanding how to maximise the reputational protection afforded by post-crisis communication (Coombs, 2007, p. 163) | Conceptual | Offline | Crisis response strategies (i.e. a manager’s denial of the connection of the crisis to the organisation, minimising the accountability of the organisation to the crisis, and bolstering strategies to rebuild the organisation’s assets) should protect the reputation of the organisation when managers use a combination of shaping attributions of the crisis, changing perceptions of the organisation in crisis and reducing the negative affect generated by the crisis Organisations should consider consumer sentiment towards a brand during a crisis and how this influences their brand attachment. Negative consumer emotions can be hard for organisations to recover from | Coombs, W.T. (2007), “Protecting organization reputations during a crisis: the development and application of situational crisis communication theory”, Corporate Reputation Review, Vol. 10 No. 3, pp. 163–176, doi: 10.1057/palgrave.crr.1550049 | |
| The Social Media Users’ Crisis Response (SMUCR) Framework | Crisis communication is crucial to successfully protect and defend a brand in a time of need (Morgan and Wilk, 2021, p. 1) | Empirical, qualitative | Offline | Social media sentiment towards a brand crisis on a global scale shifted dramatically over a 12-month period with some time periods showing that consumers are more sensitive to sentiment than others, such as immediately after the crisis while another incident was taking place, which heightens the crisis making it difficult for the brand to recover Four stages of social media user’s crisis response were found: Stage 1 of the SMUCR is “Framing and Identity Development”, Stage 2: “Sense-making”, Stage 3: “Accountability”, Stage 4: “Vigilant Rebuilding”, giving rise to the original SMUCR Framework | Morgan, A. and Wilk, V. (2021), “Social media users’ crisis response: a lexical exploration of social media content in an international sport crisis”, Public Relations Review, Vol. 47 No. 4, p. 102057, doi: 10.1016/j.pubrev.2021.102057 |
Source(s): Authors’ own creation
Methodology
Sample
As the objective of this study was to explore social media users’ responses to a crisis, it was deemed that X (Twitter) was a suitable social media platform from which data could be collected. Other studies (e.g. Wilk et al., 2022a, b) have also used Twitter to explore a new phenomenon in social media user’s narratives, as tweets are typically publicly facing and contain people’s opinions, views and/or experiences about a variety of phenomena. Barbagallo et al. (2012, p. 507) noted, that the shortness of the tweet encourages users’ to focus on the essence of the information portrayed in that message and that message is ultimately easier to consume. Therefore, publicly available social media UGC in the form of X (Twitter) posts and comments which featured the hashtag “#COVID” or “#COVID-19” and the word “supply” were harvested using the global social media monitoring platform – Salesforce Social Studio (Chae et al., 2015). These hashtags and the keywords were deemed as the best identifiers of the type of content which would be suitable for informing this study and assisting in answering the research questions (O’Hallarn et al., 2018). The time parameter used for data collection ranged from February 2020 to December 2022. This period was deemed a suitable data collection window, as it is associated with the global outbreak of COVID-19 announcement (WHO, 2020) and includes the end of the main COVID-19 wave (Maragakis, 2021). The length of time (3 years) is also sufficient as it presents a longitudinal overview of the social media narrative about COVID-19 supply chain issues. Based on Gaskell’s (2000) approach, the 3-year timeframe is sufficient as no further new insights into the crisis are expected to illuminate from the data due to thematic and data saturation.
These search parameters acted as a set of filters for identifying conversations relevant to this study. The Salesforce Social Studio platform returned the social media UGC data which included original social media (X) posts and responses (i.e. comments) together with timestamps (date and time). The social media UGC (posts and comments) were treated as observations (i.e. rows within the spreadsheet which housed the data. The data was de-identified at the collection stage and exported as an .csv file for further analysis with Leximancer. In total, there were 295,024 posts and comments collected for this study and this number encompassed all tweets relevant to the parameters set out in the data collection. Of this total, 49.4% were negative (n = 145,802), 43.4% were positive (n = 127,900) and the remaining 7.2% were neutral (n = 21,322). As in Morgan and Wilk (2021), the scoring algorithm of Salesforce Social Studio assigned a label to sentiment type, based on positive and negative evidence words and other visual cues (such as emojis and emoticons) (Salesforce, 2024). The number of posts and their respective collection period is summarised in Table 2.
Sample overview
| Sentiment | ||||||||
|---|---|---|---|---|---|---|---|---|
| Period | Negative | Neutral | Positive | Total | ||||
| Feb – May 2020 | 11,770 | (56.6%) | 355 | (1.7%) | 8,674 | (41.7%) | 20,799 | (7.0%) |
| Jun – Dec 2020 | 23,191 | (53.9%) | 8,085 | (18.8%) | 11,737 | (27.3%) | 43,013 | (14.6%) |
| Jan – May 2021 | 25,688 | (39.9%) | 3,516 | (5.5%) | 35,154 | (54.6%) | 64,358 | (21.8%) |
| Jun – Dec 2021 | 37,709 | (49.1%) | 3,062 | (4.0%) | 35,959 | (46.9%) | 76,730 | (26.0%) |
| Jan – May 2022 | 9,075 | (49.4%) | 2,436 | (13.3%) | 6,858 | (37.3%) | 18,369 | (6.2%) |
| Jun – Dec 2022 | 38,369 | (53.5%) | 3,868 | (5.4%) | 29,518 | (41.1%) | 71,755 | (24.3%) |
| Total | 145,802 | (49.4%) | 21,322 | (7.2%) | 127,900 | (43.4%) | 295,024 | (100.0%) |
Source(s): Authors’ own creation
Thematic content analysis with Leximancer
Leximancer, a text mining, qualitative data analysis program enabled by machine learning Bayesian algebra-based algorithm, was chosen for the thematic, content analysis required to answer this study’s research questions (Leximancer, 2023). The program is suitable for text analytics of large datasets of unstructured content and assists in visually displaying the conceptual structure of text (Leximancer, 2023). It enabled an automated and objective, thematic visualisation. Wilk et al. (2019) noted that their Leximancer analysis revealed insights (themes which emerged from the data) that they were not able to obtain with QSR NVivo. Other researchers have also noted that Leximancer “(makes) the analyst aware of the global context and significance of concepts and (helps) to avoid fixation on particular anecdotal evidence, which may be atypical or erroneous” (Smith and Humphreys, 2006, p. 262). Following the conventions of Leximancer use in similar types of research studies by Meek et al. (2021), Morgan and Wilk (2021) and Wilk et al. (2019, 2021, 2022a, b), this research set out to utilise the program to identify key concepts, their associations and themes. The thematic analysis enabled an exploration of social media UGC via data visualisation in a Concept Map and supported by a quantitative analysis in the Insight Dashboard report.
Results
2020 content analysis
February–May 2020
The Leximancer-driven content analysis revealed that despite the presence of dual polarity sentiments, very different themes emerged in the first time period at the outset of the pandemic (February–May 2020) in comparison to the second half of 2020 when the pandemic was quickly evolving (June-December 2020) (Figure 1).
In the February–May 2020 time period, there was evidence of immediate reaction by social media users seeking to blame a source for the pandemic and blame for the unpreparedness of the country’s response to the pandemic-affected markets. Social media users were seeking identification of the source which caused the pandemic and seeking clarity about where COVID-19 had originated from and who was responsible for the unpreparedness for this crisis. Social media users focused on identification of the crisis as #Covid or #COVID19 or #COVID-19, and the identification of the source of the outbreak (“China”) and blamed the country’s government or key figures, such as heads of state (e.g. “Barack Obama”) for the unpreparedness in responding to the crisis and managing the supply chains. For example:
Barack Obama left office in January 2017…President Donald Trump repeatedly tried to pin blame on his predecessor for the testing failures and shortage of medical supplies.
Americans blame China for the spread of the coronavirus, and yet they’re taking advantage of the crisis to make the world more dependent on their supply chains.
The UGC exhibited confusion and concerns about “supply” as an immediate response to the pandemic outbreak, evidenced by “supply” and “chain” and “supply” and “chains”. In the “supply” context, social media users commented about “shortage” of “food”, “meat” and “masks” due to “demand”. Social media users speculated about the ability to “buy”, “shortages”, “price” and “prices” of “food”, “plants”, “products”. Of most concern were “crude prices”, “meat shortages”, “masks shortage”, “medical (equipment) shortage” and “demand and prices”. For example:
Non-functioning supply chains are going to have lasting, hard-to-reverse effects on the food supply, medical equipment/drug supply, possibly fuels.
Government needs …to quickly design policies that effectively mitigate the impacts of COVID-19 that threaten to disrupt supply chains and livelihoods.
During this initial period, positivity centred around “supply” of “essential equipment” and “medical” “equipment” and “masks” and consumers offering solutions to managing supply shortages. For example:
Positivity about supplies:
A Royal Air Force… has helped deliver essential equipment to …enable the construction of an oxygen generation plant to supply the King Edward Memorial Hospital.
Defending the governments and agencies responses to the pandemic:
… efforts to control the spread of coronavirus have been significantly aided by the rapid supply of medical equipment…
Information sharing about the supply-chains affected by the pandemic:
The EU even dropped the requirement to certify certain short supply medical equipment…
The positivity seems to elicit dimensions of OBA, such as: positivity, information sharing, defence and virtual visual cues (e.g. Wilk et al., 2020, 2021).
Negativity is evidenced in blame about “demand and prices”, “masks shortage”, “meat shortage” and “crude prices”. For example:
…there also seems to be no short supply of price gougers, profiteers, counterfeit PPE sellers setting up fake websites to scam vulnerable people.
The negativity seems to elicit some dimensions of OBD, such as attacking the brand, negative brand knowledge and experience sharing and negative virtual visual cues (e.g. Wilk et al., 2022a, b).
June–December 2020
In the June-December 2020 time period, the social media narrative centred around framing and solution seeking the crisis. Negativity evidenced about “total number” of “deaths” and “cases”, as well as available “information” and people were testing “positive”. Of main concern was the supply of COVID-19 detection “tests” and the supply and use of “masks”. For example:
There is a huge surge in demand for COVID-19 tests, with patients flooding clinics…
Coronavirus #quarantine guards …were recruited via @WhatsApp, then “told to bring their own masks”.
Positivity focused on “safe” “vaccine” being available and the anticipation of availability of “vaccines”. For example:
A #coronavirus vaccine developed by the University of Oxford appears safe and trains the immune system to fight #Covid19 #coronavirus.
Britain has secured access to 90 million doses of potential #coronavirus #vaccines in deals with biotech firms BioNTech, Pfizer and Valneva.
The positivity seen in these responses seems to elicit dimensions of OBA, such as positivity, information sharing, defence and virtual visual cues (e.g. Wilk et al., 2020, 2021). Positivity also ensued through the protection of one’s self-best interest such as through “hoarding” of grocery items as a response to “limits” and low “supplies”.
Negativity evidenced in “deaths” from the “outbreak”, increasing “number” of “people” being “infected” and the quick spread around the world, leading to “lockdowns” affecting consumers being able to purchase required items, “limits” in purchases and “restrictions” in a supply-affected marketplace. Moreover, negativity concerned the response to the outbreak and supply affected consumers from “Trump” in “America”. For example:
US criticized for hoarding world's supply of #coronavirus drug #remdesivir
There have been 6 deaths in past 24 hours, all believed to be connected to an aged care setting. #6Newsau #COVID19Vic
Notably, the negativity seen in this time period related to country-specific supply chain matters and seemed to elicit some dimensions of OBD, such as: attacking the brand, negative brand knowledge and experience sharing and negative virtual visual cues (e.g. Wilk et al., 2022a, b).
2021 content analysis
January–May 2021
Social media UGC from the 2021 period revealed positive and negative sentiments about supply in the context of COVID-19 (Figure 2). This period was marked as sense making and solution seeking to the evolving pandemic, specifically, solution products such as vaccines. The first 6 months of 2021 aligned with different UGC themes in comparison to the second half of that year, suggesting social media users were communicating about different aspects of the COVID-19-affected supply chains.
In the first 6 months of 2021, marking the start of the second year of the pandemic, the focus was on the supply of and “distribution” of “vaccines” produced by “Astrazeneca”, “Pfizer”, “Moderna”, “Sinovac”.
Positive comment example:
Moderna COVID-19 vaccines extremely efficient after first shot COVID-19 vaccines …lowered the chance of an infection by 80 per cent.
Negative comment example:
Germany, France and Italy suspending Oxford-AstraZeneca coronavirus vaccine amid blood clot concerns.
Positivity existed around “federal” agreements in various countries, and the supply of the vaccines and their “availability”. For example:
In the U.S., 88 COVID-19 vaccine doses have been administered for every 100 people… The U.S. government secured major supply deals with leading vaccine makers, notably Moderna and Pfizer.
This indicated other stakeholders are able to assist with the crisis.
Positivity existed around the vaccines being “free” and their “safety”, with the focus being on each country rather than the globe, through information sharing, positive virtual cues and brand defence (Wilk et al., 2020, 2021). For example:
Vaccines have not been shown to cause blood clots…health officials have encouraged pregnant people to get vaccinated against coronavirus, saying shots are safe.
Negativity existed around “surge” in “infections”, “shortages” of “oxygen”, “oxygen tanks” in hospitals in countries like “India”, “Italy” and “Brazil”. For example:
Oxygen supply to last only half an hour now, more than 200 lives are at stake. We lost 20 people due to an oxygen shortage last night #Oxygen #coronavirus #COVID19
Negativity included dimensions of OBD, such as attacking the vaccine or hospital brand, dissuading others from going to the hospital or getting vaccinated with a particular brand and using negative (such as angry, disappointed, crying) emojis and emoticons (Wilk et al., 2022a, b).
June–December 2021
In the second half of the second year of the pandemic, social media users commented on the COVID-19 “impact” on the “industry”, “business”, “market”, “price”, “supply chain”, “disruptions”, “economic” “trade wars” and “inflation”.
Positive comment example:
Gold the precious metal remains steady above $1780 on Thursday, as worries over rising inflation and supply chain issues boosted the safe-haven appeal.
Supply chain has faced challenges in getting back up to speed following the pandemic, leading to price spikes on many products.
Negative comment:
Fresh coronavirus outbreaks are forcing factory shutdowns in countries such as Vietnam and Bangladesh, aggravating supply chain disruptions that could leave some U.S. retailers with empty shelves.
Specifically, “oil” and “energy” “supply” “disruptions” were of concern to consumers.
For example:
#Oil rose to more than $83 a barrel, supported by tight supply and expectations that rising coronavirus cases and the spread of the #Omicron variant will not derail a global demand recovery.
This suggested that the supply chains were affected by the crisis. However, positivity did exist. For example:
OPEC agreed to increase their output by 400,000 barrels per day monthly.
Negativity was also evident:
Electricity problem, Something is brewing and people are trying to brainwash us again, it is not going to happen. BTC haters everywhere, because they know what is coming.
2022 content analysis
Analysis of 2022 UGC identified positive and negative sentiments about supply in the context of COVID-19 throughout the third year of the pandemic (Figure 3). The first half of 2022 signalled vigilant rebuilding across the globe and preparedness for upcoming crises such as the “Omicron” and then the “monkeypox” in the second half of the year.
Positivity was evident in social media content about single concepts such as the availability of “masks”, “information” and “vaccinations” such as “Astrazeneca” and “Pfizer”.
Positive comments included:
India offered Bangladesh 1.2 million free doses of the AstraZeneca coronavirus vaccine.
Pfizer is preparing to ship out its supply within 24 hours.
Negative comment example:
Wealthy countries have already bought up almost the entire supply of Pfizer’s Paxlovid and Merck’s Lagevrio for 2022. The companies are gatekeeping which countries can produce the drug.
Positivity existed about “market capacity”, “work from home” products and “recovery” within the “supply chains”. For example:
Tatarstan intends to increase the export of halal products by 25–30% and reach about $ 7 million by the end of the year. The increase is attributed to the cancellation of coronavirus restrictions…
Negativity was present moreso in the second half of the year (June-December 2022) with new “variants” of COVID-19 sweeping the world, including “monkeypox” and new world political environment destabilising due to the “Russia-Ukraine” war affecting the recovering supply-chains and “manufacturing” of certain products. For example:
…uncertainty surrounding the Russia-Ukraine war and resurging COVID-19 cases in China…resurfaced the fears of further disturbances in the supply-chain distribution.
Negativity through “disruption” and “bottlenecks” in the “supply chains” such as in “energy” and “consumer-goods” was commented about by social media users who were now being affected by “inflation”, the prospect of “famine” and decreased “production”. For example:
The world is plagued right now with a supply chain crises, famine crises, war crises, coronavirus crises, financial crises, inflation crises!
Discussion
COVID-19 wreaked havoc around the world, affecting supply chains and subsequently, causing consumers to voice their frustration to their social media networks. This study extended the SMUCR Framework by Morgan and Wilk (2021) and provided insights into OBA and OBD (Wilk et al., 2019, 2020, 2021, 2022a, b, 2023, 2024) as well as into the occurrence of positivity amongst negativity by Wilk et al. (2023) and Prins and Wellman (2021). It explored social media narratives over a longitudinal, three-year period and showing that there was positivity amongst negativity throughout the COVID-19 pandemic, globally. However, despite there being similarities in sentiment across the three-year period, there were very different themes aligned with the six time blocks investigated, helping to gain a better understanding of how the SMUCR Framework evidenced itself in this context, specifically, that a nine-stage response was evident. Notably, the traits of positivity and negativity gave rise to OBA (Wilk et al., 2020, 2021) and OBD (Wilk et al., 2022a, b) at various points of the crisis response over the three-year period. Moreover, our study is the first to show the existence of sentiment coupling in social media users’ crisis responses.
There was support for Hypothesis 1, as the crisis response by social media users spans over nine stages rather than four as originally reported in Morgan and Wilk’s (2021) SMUCR Framework. It would seem that the condensed four-stage SMUCR Framework would be suitable for a one-brand, short-term crisis scenario. When the crisis spans a multitude of brands globally, over an extended period of time, such as the COVID-19 pandemic which affected government and destination brands (countries), vaccine brands, food brands, personal protective equipment (PPE) brands and a multitude of other brands across most industries and sectors, social media users seem to evidence their crisis response over nine stages as reported by our study. Thereby, this study has shown that whilst some parts of the original Morgan and Wilk’s (2021) SMUCR Framework do apply to a global crisis affecting a multitude of brands, some parts of the Framework required an adjustment as additional stages of social media users’ responses were evidenced in this study.
Hypotheses 2 and 3 were also supported by this study. Sentiment coupling was found to exist through positivity and negativity co-appearance in social media UGC during the COVID-19 which evidenced the supply chain crisis. However, as expected, not all positivity had dimensions of OBA and not all negativity was OBD, as some social media UGC was not experience-based, which was a required characteristic for OBA and OBD, and some UGC did not include brand defence, which was a requirement for OBA and was instead a mere passing on information or insights into the crisis. However, most notably, the SMUCR Framework stages do seem to incorporate traits of OBA and OBD, giving rise to a more comprehensive and robust SMUCR Framework, one which includes their existence and therefore requires their inclusion in crisis management strategies.
Evidently, at the outset of the pandemic, social media users exhibited the stage of immediate reaction and activation of their online networks in reaction to the outbreak of the global pandemic, by seeking to blame the source, seeking to identify the source which caused the pandemic and seeking clarity about where COVID-19 had originated from. Immediately, social media users gave the pandemic an online identifier as either “#Covid”, “#COVID19” or “#COVID-19”. This is in line with the first stage of the SMUCR Framework by Morgan and Wilk (2021), “Framing and Identity Development”. However, unlike Morgan and Wilk (2021), the present study found that the identification of the crisis and the identification of the source of the crisis as well as online tribe activation, came before the framing, which took place in the second half of the first year (June-December 2020). The first stage of the SMUCR was therefore extended to include online tribe activation in blaming the suspected culprit and speculation and concern (Table 3) which appeared to align with OBA (Wilk et al., 2020, 2021) and OBD (Wilk et al., 2021). This resulted in the revised Stage 1 of the SMUCR Framework (Figure 4, Table 3).
Revised social media users crisis response (SMUCR) framework
| Original SMUCR framework stages (Morgan and Wilk, 2021) | Original SMUCR framework stage descriptors (Morgan and Wilk, 2021) | Revised SMUCR framework stages (this study) | Timeframe (this study) | Revised SMUCR framework stage descriptors (this study) | OBA, OBD, eWOM |
|---|---|---|---|---|---|
| 1. Framing and identity development |
|
| Feb-May 2020 (Crisis Outbreak Globally) |
|
|
| 2. Sense making |
|
|
|
|
|
| 3. Accountability |
|
|
|
|
|
| 4. Vigilant rebuilding |
|
| Jan-Dec 2022 |
|
|
Source(s): Authors’ own creation
The second half of 2020, saw social media users blame a multitude of stakeholders for the COVID-19-affected supply chains, including governments and leaders. Social media users’ immediate reaction appeared to have been met within some hesitancy and confusion over which vaccine to trust. They engaged in OBA and OBD about potential solution brands by AstraZeneca, Pfizer, Moderna and others. This period confirmed and extended the Sense Making stage within the SMUCR Framework (Morgan and Wilk, 2021), including: 3. Framing and Blaming (Zhang et al., 2020), 4. Sense Making (Stieglitz et al., 2018), 5. Solution Seeking and 6. Solution Implementation Initiation (Figure 4, Table 3).
The second year of the pandemic, 2021, marked a period of cautious solution implementation as consumers became attuned to the crisis-affected supply chains. Specifically, social media users were engaging their networks to assess and criticise the implementation of solutions, such as which brand of vaccine is available, their side effects and which is more effective. The focus shifted to the changes occurring or being implemented by individual brands and outlets. A new pandemic in the form of “monkeypox” emerged to which social media users responded in a more calm and calculated manner. The third year confirmed the Vigilant Rebuilding stage and additionally gave rise to two new stages within the SMUCR Framework, namely: 7. Solution Implementation, 8. Vigilant Rebuilding (e.g. Frederick and Pegoraro, 2023) and 9. Preparedness for the next crisis (Figure 4, Table 3).
Each stage of the SMUCR Framework illustrated sentiment coupling, evidenced by negative and positive UGC to the COVID-19 supply chain crisis. This showed that social media users draw on their positive and negative brand perceptions, attitudes and behaviours which they share with their social networks. Specifically, where social media users did not have experience with the COVID-19-affected supply chains, they would engage in positive and/ or negative word of mouth (WOM) behaviour; whereas, where they did exhibit experience in this context, they would then engage in OBA and/or OBD, which were not mutually exclusive and did co-appear together. Notably, characteristics of OBA and OBD were evident in social media UGC (Wilk et al., 2021, 2022a, b). Consequently, in light of the scale of a global crisis, such as COVID-19, evidence from this study supports a revision to the original SMUCR Framework. Although extending the number of stages from four to nine presents a certain level of complexity, the revised framework adds more nuances and provides additional details to successfully manage a large-scale crisis. The revised SMUCR Framework is presented in Figure 4 and further explained in Table 3.
Research implications
Theoretical implications
This study enhances the SMUCR Framework by incorporating the evolving role of social media users in shaping brand narratives during crises. It emphasises the phases of user engagement and the influence of UGC on OBA and OBD. The results highlight the fluctuating nature of public sentiment, providing a detailed understanding of how consumer voices influence brand perception over time. This theoretical contribution offers a fresh perspective on crisis management, stressing the need to monitor and adapt to changing social media discourse to sustain brand resilience.
This study furthers the original work by Morgan and Wilk (2021) and updates the SMUCR Framework from four to nine social media users’ crisis response stages. It sheds new light on the role which social media users play in crisis evolution and management online (Wilk et al., 2023). It would seem that the original SMUCR Framework would be suitable for a one-brand, short-time frame crisis scenario, such as the ball-tampering Cricket Australia crisis which Morgan and Wilk (2021) reported on; whereas, the present study proposes a Framework for a crisis which is more complex in nature and spans a multitude of brands globally, lasting over a prolonged period of time. The updated nine-stage SMUCR Framework can be applied to various brands within the industry or product category, specifically in the context of a global, long-term crisis, such as COVID-19. This study presents a refreshed view of the complexity of crisis management in the digital age.
Furthermore, this study is the first to show evidence of OBA and OBD (Wilk et al., 2019, 2020, 2021, 2022a, b, 2023), and sentiment coupling within the nine-stage SMUCR Framework. It proves that consumers as active social media users are enablers of crisis response within the online communities and networks relevant to the crisis-affected brand. The study highlights that consumers and other social media users are the facilitators of positivity and negativity online, but this strong sentiment evolves over the crisis timeframe fuelled by UGC. Notably, this seems to have affected the way consumers engage in consumer purchase decision-making and consumption-related goal-setting in online platforms.
Managerial implications
This study validates the important and influential role that social media users play in crisis management. The interplay between positivity and negativity through OBA and OBD in social media UGC about a crisis has the potential to influence public opinion and affect a business and its operations. Business professionals, specifically brand and public relations managers, supply chain strategists and crisis management consultants should be aware that crisis management strategies and practices are not complete, nor will they be effective, without the involvement of social media users at every stage of the crisis response.
From a traditional marketing macro-economic perspective, marketing and management act as a subset of the political and legal environment. The two functions do not work in isolation – these are influenced by the environment and by the activities of each function. Supply-chain managers should anticipate and communicate potential delays in the supply-chain transparently to marketing and innovation teams (such as vaccine development departments) so that they can cross-collaborate across departments regarding any potential delays. Essential to this process is marketers’ transparency in communicating with stakeholders, including suppliers and their customer-base, via digital platforms (Wilk et al., 2022a, b, 2023). Such communication plays a pivotal role in managing the public’s expectations around distribution. There should be enhanced efforts to coordinate aligning brand positioning and crisis response strategies across traditional and digital channels, ensuring a proactive approach in addressing crises on social media, as social media users are important stakeholders in the supply chain process. Social media managers of vaccination companies, such as Pfizer and Moderna, should employ targeted communication strategies and authentic sources of information, that actively address the concerns inherent in UGC, e.g. for Pfizer (February-May 2020; Jan-May 2021; 2022) and Moderna (January-May 2021). If customers can identify with a company’s brand, then this ultimately serves a higher-order purpose than utilitarian value (Bhattacharya and Sen, 2003). Such managers should respond to users’ queries by proposing a solution to the crisis. This is to rebuild trust and vigilance in acknowledging the source of the blame, then moving forward with optimism (Coombs, 2007; Morgan and Wilk, 2021) in the supply-chain. This should result in users’ spreading OBA within the online networks, thus strengthening the brand’s long-term relationship with customers, brand commitment and upholding its reputation within the customer base (Bhattacharya and Sen, 2003).
Organisations should implement robust social media monitoring tools such as Salesforce (2024) and Hootsuite (2024) to track and analyse emerging consumer trends, changing sentiment, engagement and key topics of interest. Real-time monitoring and data analytics on a dashboard, which is supported by artificial intelligence (AI) practices, can help to identify disruptions to the supply-chain and further inform communications. Utilisation of such technologies can help social media managers to pinpoint details and enable them to communicate with enhanced accuracy and precision of supply-chain and market movements (Wilk et al., 2022a, b, 2023). Ensuring a cohesive response during crises entails investments in employee training and upskilling them in the use of technologies to monitor UGC. Stakeholders, such as consumers and employees, who are digitally competent can also contribute to OBA (Badrinarayanan and Sierra, 2018) rather than form a dislike towards the brand (Fetscherin, 2019), which could manifest OBD.
A holistic crisis response plan (Taylor, 2010) signifies that social media content during a crisis is multifaceted and therefore requires a rigorous response. The new SMUCR Framework should be integrated into managers’ current crisis response and management practices as well as supply chain strategies. As an example in the vaccine industry, updated strategies should target communications to current, loyal (return vaccinators) and potential customers (those in consideration of vaccines; or non-loyal brand switchers) in order to pre-empt and navigate forthcoming challenges. Focus on transparent communication and information dissemination during crises to address concerns and provide accurate information to the public (Nayate et al., 2004) is imperative so that users will feel reassured once supply-chain managers proceed in implementing a solution.
Brand and public relations managers can leverage insights from this study to better understand the stages of social media user engagement during crises, allowing them to craft more effective communication strategies that foster OBA which includes positivity, brand defence and in-depth brand information sharing and mitigate OBD. This study presents further evidence of the importance of social media users in brand management online and encourages brand and public relations managers, specifically marketers, to ensure their crisis management strategies include the brand’s online community of social media users in fuelling OBA at a time of a crisis to ensure brand survival. As an example, GameStop would be on the brink of bankruptcy had it not been for its loyal online brand advocates, that is their loyal and committed customers, who rallied around the brand in this crisis and fundraised money to rescue the brand (Laughlin, 2024).
Additionally, supply chain strategists are able to use this study’s insights to more accurately anticipate and manage the impact of supply-chain disruptions on brand perception, ensuring more resilient supply chain practices through the involvement of social listening and social media user engagement. This is not a typical practice in the supply chain management context and one which offers much merit for more effective supply chain practices in a time of crisis (Wilk et al., 2023, 2024).
Future research
In terms of future research, the updated SMUCR Framework should be tested in other crises and consumer behaviour contexts to prove its validity. Furthermore, at a contextual level, future studies could investigate whether the order of the additional stages of the SMUCR Framework differs depending upon their context of study; and whether the co-occurrence of positivity and negativity both inherent in OBA and OBD also shift in line with the context. The key differentiating factors between OBA and OBD are not only the sentiment type (or valence); more importantly, OBA and OBD are different to electronic WOM (eWOM), as they are experienced-based and more intense in expression than eWOM (Wilk et al., 2018, 2021, 2022a, b). For example, in the COVID-19 context, Aksoy et al. (2023) show how in an organisational setting, the spreading of positivity and negativity inherent in principles, such as organisational ethics and justice, manifest in eWOM. In that study, such manifestation of users’ opinions relating to government officials, vaccine efficacy (AstraZeneca) and supply, would either lead them toward or against a particular vaccine. Notably, future studies should investigate the phenomena of OBA, OBD and eWOM in crises and engage in quantitative research.
The updated nine-stage SMUCR Framework provides significant contributions to both academic literature and practical applications in brand management and social media marketing communication during crises. By capturing the evolving dynamics of social media user behaviour over a prolonged global crisis, the Framework enables practitioners to anticipate and respond to shifts in public sentiment more effectively. Specifically, businesses can leverage insights from the Framework to improve crisis response strategies, such as crafting targeted communication that fosters OBA while mitigating OBD. Further, supply chain strategists can use these insights to address reputational risks by transparently managing disruptions and engaging with social media narratives in real time. Ultimately, the Framework equips brand managers and crisis consultants with actionable tools to harness UGC as a driver for building resilience and strengthening brand loyalty, ensuring sustained brand equity during and after crises.
While the updated SMUCR Framework provides robust insights into social media user behaviour during a prolonged global health crisis, its applicability to other crisis contexts remains an area for future exploration. The findings of this study are grounded in the COVID-19 pandemic, which can differ when applied to short-term, cultural- or geographic-specific crises such as political conflicts.
This study also is limited to Twitter (X) data. Therefore, this study acknowledges potential biases related to platform-specific demographics and regional variances should other platforms are used. These considerations highlight the need for further research to validate the SMUCR Framework across diverse crises and platforms, ensuring its broader applicability and relevance.
Therefore, this paper encourages future studies to monitor conversations over an extended evaluation window including pre-COVID, during COVID and post-COVID. This complete event study approach provides an additional comparative dimension to measure the magnitude and dynamics of the global shock to the supply chain. We anticipate such an approach can yield interesting insights into the temporal evolution of public sentiment and discourse, particularly in relation to major disruptions.
Future research can also delve into the prolonged effects of social media narratives on brand reputation following a crisis, focusing on how consumer reactions change over time. Examining the influence of key opinion leaders in moulding public sentiment during crises and comparing their impact to that of regular users, could yield valuable insights. Furthermore, studying cross-cultural variations in social media responses to supply-chain disruptions may uncover distinct patterns and strategies for international brands. Lastly, creating predictive models with sentiment analysis to forecast changes in public opinion could refine the updated SMUCR Framework, supporting brands in proactive crisis management.
Conclusion
Social media users play an important role in the narratives about brands during a crisis. This study showed that there are several stages which social media users go through when sharing about a supply-chains crisis and its effect on brands. Moreover, consumers’ voices within their UGC are effective at spreading OBA and OBD during a crisis, which can further fuel positivity and negativity within online networks. The sentiment polarity presents valuable insights into the dynamics of public discourse via reactions during a time of crisis. The evidence from social media demonstrates non-stationary feelings expressed by the general public over time, providing strong support for an updated SMUCR Framework which offers a new lens through which companies should manage a crisis to ensure resilience in a time of need.
About the authors
Funding: The research team would like to acknowledge and thank the Edith Cowan University Early-Mid Career Researcher Pilot Project Grant 2022 for funding this research.
