چگونه نام تجاری مرتبط با محتوای تولید شده توسط کاربر در سراسر یوتیوب، فیس بوک، و توییتر متفاوت است؟

This study tests hypotheses regarding differences in brand-related user-generated content (UGC) between Twitter (a microblogging site), Facebook (a social network) and YouTube (a content community). It tests them using data from a content analysis of 600 UGC posts for two retail-apparel brands (Lululemon and American Apparel), which differ in the extent to which they manage social media proactively. Comparisons are drawn across six dimensions of UGC; the dimensions were drawn from a priori reading and an inductive analysis of brand-related UGC. This research provides a general framework for comparing brand-related UGC, and helps us to better understand how particular social media channels and marketing strategies may influence consumer-produced brand communications.

Social media have migrated into the ‘mainstream’ and marketers have taken notice: the percentage of companies using social media for marketing is expected to reach 88% by 2012, up from 42% in 2008 (Williamson 2010). Companies are leveraging social media not only for digital advertising and promotions, but also to handle customer service issues, mine innovation ideas, and ‘authentically’ engage with customers (Solis 2010). There is considerable diversity across the types of social media, which encompasses formats such as blogs, social networking sites and content communities (c.f. Kaplan and Haenlein 2010). While marketing scholars have studied certain social media channels in isolation, few have incorporated multiple types into a single study for comparative purposes. Research that compares how users engage with different social media can enhance our understanding of the variability between media, and provide managers with insights on how to allocate resources across platforms. One form of consumer engagement that can be compared across social media sites is user-generated content (UGC). User-generated content is an important means through which consumers express themselves and communicate with others online (Boyd and Ellison 2008); it is what is produced in the moment of being social, as well as the object around which sociality occurs. UGC takes on many different forms, such as Twitter tweets, Facebook status updates, and videos on YouTube, as well as consumer-produced product reviews and advertisements (c.f. Dhar and Chang, 2009 and Muñiz and Schau, 2007). Importantly for marketers, much UGC across various media is brand-related and has the potential to shape consumer brand perceptions. Casual observation suggests there is a tremendous assortment of brand-related UGC across the different social media types; for example, a YouTube video does not look like a Facebook wall post. Better understanding these differences is potentially important for marketers who are concerned with the co-creation of their brands in different social media platforms. As such, this research addresses the following questions: How does brand-related UGC vary across different social media types? And how does the extent to which social media marketing is proactively managed relate to differences across social media types? These questions have not yet been broached in the still nascent (Burmann 2010) research stream on brand-related UGC. To answer them, we draw on brand-related UGC from three different types of social media: Facebook (a social network), Twitter (a microblogging application), and YouTube (a content community). We test hypotheses using data derived from a content analysis of materials on two brands (American Apparel and Lululemon) on each site. Our study builds on previous brand-related UGC research to make three primary contributions. First, it provides a preliminary framework for comparing the content that consumers create when they produce brand-related posts. Second, it improves our understanding of how specific social media channels influence the brand-related messages that consumers create (c.f. McLuhan 1964). Third, it highlights differences in UGC where social media is more, versus less, proactively managed by brands. Thus, it can help inform managerial decision making about social media. The paper proceeds as follows. It briefly reviews the literature on UGC and the social media sites of interest. It then develops research hypotheses and describes our methods. Finally, it presents results of the content analysis and a discussion of findings.

This study offers three contributions. First, it provides conceptual insights into how different social media sites foster UGC with different characteristics. Second, it develops some preliminary ideas on how proactive social media marketing relates to UGC. Third, it offers a preliminary set of dimensions for comparing brand-related UGC found on different sites. Our findings show that while brand-related UGC tends to differ across sites for some facets of content (particularly promotional self-presentation and brand centrality), it does not do so for others. In addition, some cross-site content patterns (e.g., ‘marketer-directed communication’) appear to differ significantly for brands that are more, versus less, proactively managed. The strongest site influences in brand-related UGC seem to stem from YouTube's culture of self-promotion. True to its tagline, YouTube is focused on broadcasting the self. While brands may play a role in consumers’ presentation of face (Goffman 1959) on YouTube, it appears to be a supporting rather than a central role. This does not, however, mean that brand-related YouTube UGC is devoid of brand information: YouTube can provide factual information about brands, albeit often peripheral to the main messages that posts convey. Also, for brands like AA that face criticism, the site is somewhat of a haven relative to Facebook and Twitter, which feature more negative sentiment, perhaps because their cultures support conversation and link sharing. Twitter is most distinctive from YouTube. Brand-related UGC on Twitter is least likely to feature consumer self-promotion; people more often use it to engage in discussions and spread news. Correspondingly, brand centrality trends highest for UGC on Twitter. Thus, Twitter is potentially both a boon and a bane for marketers. For proactive brands like LLL, Twitter users appear willing to use the channel to communicate with the marketer, enabling them to bask in positive sentiment and respond to negative posts. For brands like AA that are less proactive in social media, consumers will infrequently initiate marketer-directed content, presumably because they have little reason to expect it will be acknowledged. Furthermore, Twitter appears to be a particular point of vulnerability for brands receiving critical attention, yielding less positive content, and more neutral and negative content (for a brand like AA). Facebook, with its myriad of UGC types, seems to fall somewhere in between YouTube and Twitter. Consumer self-promotion features more prominently on Facebook than on Twitter, but less so than on YouTube. UGC that highlights the brand centrally is also more likely on Facebook than on YouTube. As with Twitter, it appears that proactive marketer attention is required if marketer-directed content and response to marketer actions are desired; such content seems to be higher on Facebook for a proactively managed brand like LLL than for one like AA. Though tentative, our findings have implications for marketers investing in social media. Particular media, such as Twitter and Facebook, seem to offer more opportunities for marketers to collaborate with consumers to circulate positive sentiment about, and increase the visibility of, brands. This appears to be enhanced by a proactive social marketing strategy: brands should provide a space where conversation can occur, not only with them, but also with other consumers (c.f. Muñiz and Schau 2011). Marketers should also ‘entice’ consumers to participate, through relevant and valuable content, as well as to ‘validate’ their participation by responding to them. Sites such as Twitter may be a source of both threats and opportunities for brands experiencing unfavourable exposure. Monitoring the site and addressing potentially problematic posts may be one way marketers can dampen budding issues; as such, being active in providing information, listening, and participating in the social give-and-take on Twitter may be a priority for marketers. At the same time, ignoring YouTube may be a wasted opportunity for marketers. In particular, those seeking a subtle life-world placement and association with a particular constellation of brands might be well-advised to become actively engaged with YouTube. Future research on brand-related UGC is required both to confirm and extend the tentative insights we advance here. Studies could analyze additional types of social media (e.g., collaborative projects, virtual social worlds, and virtual game worlds), which seem to vary considerably from the ones examined, as well as other content dimensions (e.g., response to other consumers, requests for information, etc.) that might be actionable and relevant for marketers. Analysis could consider user's perceptions of their audience across different social media categories and how that might influence what users post (thanks to a reviewer for this suggestion). Extensions could examine how some of the content dimensions associate with traditional metrics that are important to marketers, such as sales, as well as with others that are becoming more important to social media marketers (c.f. Hoffman and Fodor 2010). Work could seek to understand how some of these dimensions relate to consumer meanings derived from the UGC (e.g., how do derived consumer meanings differ when one brand is central vs. when a brand is peripherally included in a constellation of similar brands?). Future research could also work to address limitations of this study, such as its narrow scope on one category. We close by acknowledging other limitations of this research. One relates to how the data was gathered for the study. The data was harvested using Google search results between December 2010 and January 2011. Google's results draw on Twitter's archive, but Twitter deletes the number of tweets it publically archives over time (as does Facebook). For example, on August 1 2010, Twitter may have archived 500 tweets from July 1, 2010, but by December 1, 2010, there may only be 200 tweets archived for July 1. There is no way of knowing whether or not this deletion is systematically biasing the sample of Tweets (and Facebook posts) collected. This suggests that the results from this study should only be extrapolated beyond this sample with extreme caution. Further reinforcing this caveat is the fact that this UGC sample draws from only two brands, both targeting mostly female consumers. As such, most of the UGC for both brands was created by females, and research with more diverse samples is clearly warranted. In addition, one of the two brands studied was under negative scrutiny for numerous reasons, which may limit the generalizability of the pattern of findings associated with it. Along these lines, research focusing on different types of brands (e.g., brands more closely tied to one's self-concept), representing a broader variety of product/service categories, would also be valuable. Some types of brands might be more likely to elicit a wider range of UGC (e.g., entertainment-related brands that inspire fan fiction); others, even those generally liked by consumers, might sponsor a more limited range (e.g., brands targeted toward businesses). Based on the findings from this study, we would expect some patterns of results (e.g., those relating to brand centrality or factuality) to be less variable across brands, while others (e.g., those relating to marketer-directed communication or brand sentiment) might be more variable across brands. Nevertheless, this study raises a useful set of implications for managers and provides a number of possibilities for future research.