Foodora and UberEats are both brand new companies and therefore little research is available analyzing the Twitter interactions of these two companies. This review will therefore delve into the literature focusing on the most thorough and resourceful means that have been explored to analyze the reach of a Twitter platform. This review will explore the special difficulties that a brand Twitter faces and what certain brands have done to become popular on Twitter despite these difficulties. This review will also look into the methods that a Twitter page can use in order to reach a wider audience. This information will help to determine how best to interpret the Twitter data for Foodora and UberEats and to determine what further courses of action these two companies can take in order to improve their reach on Twitter.
When a Twitter network is analyzed, the first problem is to identify the scope and shape of the network. According to a consensus of sources, the number of followers a Twitter page has, also referred to as its indegree, is not sufficient to accurately determine a Twitter user’s popularity, and can in fact be misleading (Cha, 2010, p. 11). This is due largely to the uneven amount of activity generated by each individual user, and their inclination to Tweet about one topic rather than another. It is therefore not sufficient to rely on the number of followers in order to determine the reach of a Twitter campaign.
Use your promo and get a custom paper on
"Foodora and UberEats: Literature Review".
As an alternative to indegree, Cha has proposed a system of popularity calculation which focuses on retweets and mentions. Cha has discovered that while a Twitter user with many followers will generate more one-on-one reactions, a Twitter users with many retweets and mentions will reach more users overall (Cha, 2010, p. 13).
When analyzing the success of a campaign, a wide variety of literary sources recommend mapping the network in NodeXL. Smith proposes that there are six structures of networks which include 2 polarized groups, 1 tight crowd, several community clusters (2014, p. 8). A brand likely has a support network type network structure in which users contact the brand and the brand tweets back, but the marketing potential of this design is limited due to the lack of outward interaction. A consensus between Smith and Kleine-Kalmer indicates that a brand will be more successful by fostering a community.
The special problem that brand Twitter pages face, as explored by Kleine-Kalmer, is that user interaction with brand Twitter pages is rare. As is reported in the book, 98.7% of the followers of a brand do not interact with it (p. 9). The literature further explains that some successful brand Twitters are able to foster a sense of community with their users, leading to more interaction between the brand and its users and even amongst the community of users of the brand (Kleine-Kalmer, p. 37). Users develop a feeling of belonging by interacting personally with a social media specialist operating the Twitter page (Kleine-Kalmer, p. 39). Users also develop a sense of a bond through participating in online events, helping to create something for the brand, and sometimes providing personal information when they feel safe doing so (Kleine-Kalmer, p. 39). These are the options that must be explored by a brand Twitter page looking to increase the scope of its marketing campaigns by creating a community-type network shape.
A consensus of recent literature has determined that a successful Twitter community, or a community that generates a large quantity of retweets, mentions, and links, has a certain shape. In the very center of the cluster is an individual or a few individuals that are retweeted and mentioned the most (Cha, 2010, p. 11). Around these individuals is a “cohesive subgroup” or subgroups which share mostly the same links and hashtags, and are the most active within the group (Hogan, 2007, p. 6). These individuals and subgroups will have special interests and topics of expertise which they are more likely to Tweet about than other topics (Eftekhar, 2013, p. 80-81). If the NodeXL model of the network matches that of a community, the first step for a brand to benefit from this is to identify the most influential people in the community as well as the cohesive subgroups. The second step is to determine which topics are the most interesting to these users. The third step is to target these users with future Tweets using the information gathered.
A second technique for magnifying the reach of a tweet has been explored by the work of Cha following an investigation into a well-established theory on Twitter network structure. This theory indicates that only certain, well-known and well-established Twitter users hold influence on the platform (Cha, 2010, p. 10). Provided this information, Cha argues that a tweet is better aimed at a single highly influential user than at a massive group of ordinary users (2010, p. 15). It stands to reason that if a brand draws a famous Twitter user into conversation, the retweets and mentions generated through the interaction will be greater than if the brand targets its own, smaller follower base.
A third technique for improving Twitter reach as identified in the literature is to alter the voice and content of the Tweets to follow the actions of the most popular pages. According to Cha, Twitter users with the most indegree are mostly celebrities as well as famous news sources but the most frequently retweeted Twitter users rarely post original content but instead share attention-worthy information as it comes to light (Cha, 2010, p. 13). A brand may meet great success by following these parameters and post information that is both topical and interesting.
This literature review has determined that a Twitter network can be mapped with NodeXL in order to identify its community elements and then foster that community both through building a sense of belonging and by identifying the interests of the most powerful elements of that community. It has also identified that retweets and mentions have more reach than indegree, and engaging with influential players is more successful than engaging with the crowd at large. This information can be used to interpret and guide the data from Foodora and UberEats.
- Cha, Meeyoung, Haddadi, Hamed, Benevenuto, Fabricio, and Gummadi, Krishna P. (2010). Measuring User Influence in Twitter: The Million Follower Fallacy. Association for the Advacnement of Artificial Intelligence. Retrieved from https://www.aaai.org/ocs/index.php/ICWSM/ICWSM10/paper/viewFile/1538/1826,2011
- Eftekhar, Milad, and Koudas, Nick. (2013). Some Research Opportunities on Twitter Advertising. Data Engineering, 36. Retrieved from https://pdfs.semanticscholar.org/7120/3854cec08f99862c412069053d92dddb7c36.pdf#page=79
- Hogan, Bernie. (2007). Analyzing Social Networks Via the Internet. Retrieved from https://pdfs.semanticscholar.org/be39/06ca5bfc196581aeaa957cc9287179819bc1.pdf
- Jansen, Bernard J., and Zhang, Mimi. (2009). Twitter Power: Tweets as Electronic Word of Mouth. Wiley InterScience. Doi: 10.1002/asi.21149
- Kleine-Kalmer, Barbara. Brand Page Attachment: An Empirical Study on Facebook Users’ Attachment to Brand Pages. Available from http://www.springer.com/gb/book/9783658124380
- Smith, Marc A., Rainie, Lee, Himelboim, Itai, and Shneiderman, Ben. (2014). Mapping Twitter Topic Networks: From Polarized Crowds to Community Clusters. Pew Research Center. Retrieved from www.pewinternet.org/2014/02/20/mapping-twitter-topic-networks-from-polarized-crowds-to-community-clusters/
