Franz-Josef Friedrich Maximilian Schrepf
10621644
Bachelor Thesis
BSc in Economics and Business, Business Studies
Faculty of Economics and Business
Supervisor:
Dennis Stevens
29
thJune, 2016
ALL ABOUT THAT BUZZ?
A STUDY ON
THE MEDIATING EFFECT OF BUZZ ON THE
RELATION OF INNOVATION AND SALES DURING NEW VEHICLE
INTRODUCTIONS IN THE U.S. ELECTRIC VEHICLE MARKET
1
ABSTRACT
Word-of-Mouth marketing and the marketing buzz can have enormous potential in raising awareness regarding a product introduction and persuading consumers into consuming these products. However it still remains unclear which factors exactly determine the buzz creating potential of a certain product. Apart from a superior marketing strategy, innovation is one of the most frequently mentioned factors facilitating the creation of a marketing buzz surrounding a given product’s market introduction. This paper seeks out to determine the extent to which innovation affects the creation of a marketing buzz and influences product sales by analyzing introductions of new plug-in electric vehicles into the United States market between January 2014 and April 2016. In order to conduct this research, extensive amounts of vehicle sales data as well as data on the consumer engagement with the Twitter accounts of the analyzed companies, a proxy for buzz, was collected and analyzed for the given period of time. In the findings of this research, consumer engagement is identified as a partially moderating variable between the independent variable innovation and the dependent variable vehicle sales. In the plug-in electric vehicle market, new vehicles which display a high degree of innovation tend to create a larger buzz surrounding their introduction and consequently sell more units than trend following vehicles. Even though the entire spectrum of factors influencing the creation of a marketing buzz remains unclear, this research supports the belief that innovation is an essential factor in buzz creation, at least in the plug-in electric vehicle market. The results shed further light on the subject of buzz creation during new product introduction, providing valuable insights and implications for future research in order advance the field.
2
Statement of Originality
This document is written by the student Franz-Josef Schrepf, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Table of Contents
1. Introduction ... 4 2. Literature review ... 6 2.1 Word-of-Mouth (WOM)... 6 2.2 Marketing Buzz ... 7 3. Conceptual framework ... 8 3.1 Consumer Engagement... 8 3.2 Innovation ... 103.3 Visualization of the Conceptual Framework ... 12
4. Methodology ... 13 4.1 Data Collection ... 13 4.2 Data Analysis ... 14 5. Results ... 16 5.1 Descriptive Statistics ... 16 5.1.1 Innovation ... 16 5.1.2 Product Sales ... 18 5.1.3 Consumer Engagement ... 20 5.2 Hypotheses Testing ... 23
5.2.1 Correlation between Innovation and Vehicle Sales ... 23
5.2.2 Correlation between Innovation and Consumer Engagement ... 25
5.2.3 Correlation between Consumer Engagement and Product Sales ... 29
5.2.4 Mediating Effect of Consumer Engagement on Innovation and Vehicle Sales ... 32
6. Discussion ... 33
7. Limitations and Suggestions for Future Research ... 36
8. Conclusion ... 38
Acknowledgement ... 39
Appendix ... 40
4
1. Introduction
Electric vehicles are not an entirely new invention. In fact, the first electric vehicles, simultaneously developed by different innovators in Hungary, the Netherlands, and the United States in the early 1800s, reached the public prior to the first gasoline powered vehicles (United
States Department of Energy, 2014). It was not until 29th January, 1886, when the German
inventor Carl Benz applied for a patent for his novel vehicle powered by a gas engine, that the gasoline powered automobile as we know it was born (Daimler, 2016). Other than gasoline powered vehicles, electric cars functioned without the manual efforts required to steer a gasoline powered vehicle, such as shifting gears, and were not reliant on an elaborate network of gas stations as they could be conveniently charged at any electricity socket. Around the turn of the century, the early 1900s, the heyday of the electric car, electric vehicles accounted for nearly one third of all vehicles on American roads. Even Ferdinand Porsche, founder of the sports car company bearing his name, developed an electric vehicle called the P1 in 1898, and proceeded to invent the world’s first hybrid electric vehicle, featuring an electric as well as gasoline engine (United States Department of Energy, 2014).
However, due to Henry Ford’s advances in the mass-production of his gasoline powered Model T vehicle, the significant financial advantage of gasoline powered vehicles over their electric counterparts, and their nearly unlimited range thanks to the discovery of the cheap Texas crude oil, sales of electric vehicles declined and the technology shifted into oblivion. It was not until almost a century later that the electric car experienced its renaissance as increasing concerns about resource depletion, sustainability, and environmental pollution are dominating the public discourse (United States Department of Energy, 2014).
While electric vehicles tended to be rather unappealing due to their usually insipid design and a lack of recharging infrastructure, over the past years, some companies have succeeded turning them into the current public object of desire. The front runner of this trend is Tesla Motors, one of the most innovative car makers worldwide, focusing solely on electric
cars. On the 31st March 2016, the introduction of the new Tesla Model 3 wrote automobile
history. On Thursday morning lines of customers could be found in front of Tesla stores around the globe, waiting to preorder a car that was only going to be released in the evening of the same day. Before the release, Tesla already counted 134.000 reservations for the Model 3, and by Saturday, Tesla Chairman Elon Musk announced a total of 253.000 reservations have been made. Tesla achieved to generate such a buzz around the launch of the Model 3 that twice as many reservations were made in its first days than the total sales of all previous Tesla automobiles combined (Bloomberg, 2016).
5
The reasons for the creation of such a buzz, however, often remain unclear. Mourdoukoutas & Siomkos (2009) state that one of the key principles for the creation of a buzz surrounding a product introduction like the Tesla Model 3 is the need to be innovative. The Tesla Model 3 combines all the technological advantages, such as a wider range through superior electricity storing technology or the super charger port, which allow to charge the vehicle’s battery up to 80% within 30 minutes, while offering the model at a starting price of 35.000$, turning it into the electric vehicle with the best price-performance ratio on the market (Tesla, 2016).
Dye (1999), on the other hand, claims that the actual product and innovation it brings is less important to the creation of a buzz rather than a superior marketing strategy aiming to stimulate conversation between consumers. According to Dye (1999) even products as generic as pharmaceutical products can create a buzz - if implemented correctly.
In order to analyze whether or not the type of innovation significantly impacts the creation of a buzz around the introduction of a new electric vehicles, researchers need the means to quantify the extent of a buzz. The introduction of social media platforms enables researchers for the first time to collect the data necessary to quantify the buzz around a new product, using likes, comments, and shares of company related content as a proxy for electronic word-of-mouth and buzz.
Therefore, it is the purpose of this case study to determine the extent to which a social media buzz mediates the relation of innovation and sales during the introduction of new electric vehicles.
The answer to this question is derived through a multi-method research approach. Firstly, the degree of innovation is determined through a qualitative analysis of the companies’ press releases regarding the introduction of the new plug-in electric vehicle. Secondly, monthly sales data of the new vehicles during their introduction and overall are retrieved from the website “insideevs.com” in order to observe the relation between innovation and sales. As companies tend to launch new vehicles irregularly throughout the globe, it was decided to only utilize sales data of the United States electric vehicle market. Its position as the largest global market allows the use of large quantities of data without dispersing the product introduction through longer time periods. Subsequently, data on the companies’ Twitter activities considering the U.S. market is collected and analyzed for the time period of 2014 until April 2016. Finally, tweets during the introduction will be analyzed to ensure that their content is related to the introduction. Through analyzing the individual factors in this multi-case study
6
the relation between buzz and sales, innovation and buzz, and the mediating effect of buzz on the relation between innovation and sales is observed and the hypotheses are tested.
2. Literature review
In order to answer the earlier proposed research question, it is the purpose of this section to outline the underlying concepts and premises of Word-of-Mouth marketing (WOM), electronic WOM, and its accelerated form, the marketing buzz. Since the marketing buzz finds its foundation in the WOM, it is crucial to discuss the basic principles and past research in the field prior to the establishment of the actual research hypotheses in the following section.
2.1 Word-of-Mouth (WOM)
Word-of-mouth can be defined as the informal passing of information through one consumer to another (Lang & Lawson, 2013). Rather than investing in company-to-consumer interaction, word-of-mouth advertising seeks out to stimulate conversation between consumers and their family members, friends, and acquaintances, creating a chain reaction of communication that could spread the company’s message throughout the entire market (Mason, 2008; Gupta & Harris, 2010; Lang & Lawson, 2013). The strength of WOM lies within the assumptions made by the information recipient. As the individuals engaging in the interpersonal informational exchange are familiar with each other, the recipient of information has an inherent belief in the motivation of the information provider. Hence, the opinion of the individual is perceived as more trust-worthy than corporate messages, assuming it is based on either perceived similarities between each other or due to perceived product or service specific knowledge (Gupta & Harris, 2010; Coulter & Roggeven, 2012).
According to Rosen (2000), purchasing is a social process rather than a sole interaction between a company and consumers, which is influenced by a variety of exchanges and influences of the social environment of the consumer. The invisible network among consumers is composed of three key elements: Hubs, in the literature described as viral mavens, individuals positioned in a central role and most promising to influence others opinion. Clusters, areas of dense connection between consumers. And connections between clusters, individuals with multiple cluster belongings which increase the reach of WOM messages into other clusters (Rosen, 2000).
In his book The Secret of Word of Mouth Marketing, George Silverman characterizes WOM as “the oldest, newest marketing medium” (2005, as cited in Kimmel & Kitchen, 2014,
7
p. 5), and despite the grammatical awkwardness, the statement perfectly defines the role of WOM in the current state of the field of marketing (Kimmel & Kitchen, 2014). The rise of the internet provided new means of interpersonal communication and facilitated a renaissance of WOM marketing, the electronic Word-of-Mouth. Through the internet, engaging in e-WOM with other individuals has become significantly easier and the range of such messages grew exponentially. Furthermore, the possibilities of content have shifted, making e-WOM ever more effective. E-WOM messages may include videos, pictures, and/or verbal content, content too detailed for traditional WOM (Phelps, Lewis, Mobilio, Perry, & Raman, 2004). Successful WOM or e-WOM marketing can result in communicational chain reactions, which can gain momentum and reach what Gladwell (2000) describes as the “tipping point”, the moment of reaching critical mass, and transcend into what marketers call a marketing “buzz”.
2.2 Marketing Buzz
In order for WOM messages to reach the “tipping point” and transcend into a buzz, as described earlier, Gladwell (2000) defined three crucial factors: The support of influencer or viral mavens, a “stickiness factor”, namely the attribute of making a message memorable, and environmental circumstances that encourage a viral message to be unleashed, such as an upcoming event.
Dye (1999) describes a buzz as “explosive self-generating demand” (p.140), characterizing the key difference between WOM and a buzz, the factor of time. While WOM marketing is a continuous process characterized through interpersonal communication with regard to a certain company, a buzz is a social phenomenon where WOM gains critical mass and “explosive” growth, mostly revolving around a single product or an event, such as a new product introduction (Notarantonio & Quigley, 2009). Buzz has the ability to come seemingly out of nowhere and transform what otherwise would have been a niche product into a mass phenomenon (Notarantonio & Quigley, 2009). In other words, buzz marketing amplifies original marketing efforts by magnifying WOM activities regarding a specific moment in time (Thomas, 2004).
One powerful example of the effect of buzz on product sales is the research conducted by Liu (2006) regarding the effect of consumer buzz on box office sales of movies. The author used the comments left on the website “Yahoo Movies” as a proxy for word-of-mouth and established a model which allows to predict box office sales based on the word-of-mouth surrounding the movie. Moreover, the author discovered that WOM activities were the most
8
active during a movie’s prerelease and initial launch week. This valuable insight will be considered in the later analysis.
The power of the online and social media buzz, however, can turn against a company if not managed correctly. To put it in the words of Jeff Bezos, the CEO of Amazon.com: “If you make customers unhappy in the physical world, they might each tell 6 friends. If you make customers unhappy on the Internet, they can each tell 6,000 friends” (as cited in Luo & Zhang, 2013, p. 214). The marketing buzz is one of the most powerful weapons of marketers in their fight for consumers’ attention. Those who nurture the WOM wisely and know how to utilize the buzz are rewarded with higher margins and the admiration of their competitors, while those who use it ill-considered are crushed by its momentum.
3. Conceptual framework
In the previous section it was discussed how Word-of-Mouth marketing and a marketing buzz associated with a new product’s introduction can be useful tools for marketers to spread persuasive content messages throughout the consumer population if used correctly. However, in order to analyze the mediating effect of the marketing buzz on the relation between vehicle sales and vehicle innovativeness, it is necessary to discuss the attributes of its proxy, the consumer engagement on social media, as well as the definition of innovativeness. This section demonstrates how consumer engagement is measured through social media, defines and describes innovativeness, and establishes the corresponding hypotheses that will be tested to answer the proposed research question.
3.1 Consumer Engagement
As competition is intensifying in the global markets of today, it is no longer sufficient for marketers to focus on solely seeking the consumers’ attention. Retaining customers and building enduring relationships with consumers, organizations, and constituents dominate the new marketing paradigm with a constant focus on increasing the overall customer life time value (Coulter & Roggeven, 2012). Especially in the online space marketers typically created product-related pages and pursued to foster customer traffic on their pages. Even though the pages did aid the company in stimulating traffic and WOM, interaction and engagement with consumers proofed to be less fluid and a company’s ability and reach of WOM was often difficult to measure. Through the rise of social media pages companies were given the necessary tools to interact with consumer and stimulated word-of-mouth, and were able to
9
accurately measure consumers’ active participation and engagement with a company through likes, comments, and shares of the company’s content. These measures are important to marketers as the source expertise or credibility of a message significantly influences the belief recipients hold regarding its utilitarian values and determines the effectiveness of persuasive communication (Coulter & Roggeven, 2012).
The advantage of social media, in this case, is not only the direct interaction of companies with consumers, but the interaction of consumers among each other’s. An individual’s likes, comments, and shares/retweets regarding company messages have a significantly higher impact on the members of the individual’s network than direct company communication. Here the individual acts as a brand ambassador, as its own reputation and familiarity with the other members of its network provide the company message with legitimacy and credibility (Gupta & Harris, 2010). Furthermore, a public display of a company’s list of followers further enhances credibility of its marketing messages as a large number of previously-enrolled network members again signals more credibility which affects consumer knowledge and liking regarding the brand (Coulter & Roggeven, 2012). The amount of pre-enrolled network members is hence a control variable that is accounted for during this case study.
The social media landscape is dominated by the two most widely used social media channels, social networks and micro blogs, with Facebook and Twitter leading the global market, respectively (Ashley & Tuten, 2015). Social networks, such as Facebook, revolve around the principle of bi-directional communication, hence direct interaction between all participating parties, whereas micro blogs like Twitter are based on the idea of uni-directional and public communication. Users of microblogs can therefore not communicate in a private, bi-directional manner through messages but solely through the use of public posts, likes, and shares.
In their research, Coulter & Roggeven (2012) discovered that there is a significant difference in the effectiveness of WOM messages in the two different systems. The researchers discovered that if individuals engage in bi-directional communication, the impact of source closeness becomes insignificant as consumers are able to directly question persuasive company-created content. On the other hand, users of the micro blog Twitter do not have the ability to question message content directly; hence source closeness plays a significant role in the establishment of credibility of persuasive messages. For this reason, this case study will focus solely on automobile companies’ activities and the responding buzz on Twitter, as the
10
uni-directional communication allows a more accurate data collection and analysis of e-WOM regarding the company and new vehicle introduction.
The data collected on Twitter is publically available and has been used in several research studies. These studies have proven its predictive abilities and therefore it can be considered a reliable source of information. For example, Luo & Zhang (2013) conducted a study regarding overall company buzz on Twitter and discovered a significant correlation between consumer engagement on Twitter and a company’s stock market performance. Lassen, Madsen, & Vatrapu (2014) even succeeded in establishing an algorithm that is able to predict IPhone sales twenty days in advance, entirely based on consumer engagement on Twitter, with an error rate similar to the one of established investment bank sales prediction tools. The authors argue that likes, comments, and retweets can be seen as a proxy for the consumer’s journey along the different stages of the AIDA model (Attention, Interest, Desire, and Action) and praised social media analytics as the modern approach to sales prediction due to its ease of data collection.
In conclusion, it is expected that the following hypotheses hold true:
H1a: Consumer engagement is positively correlated to sales during the vehicle launch
H1b: Consumer engagement is positively correlated to overall vehicle sales
3.2 Innovation
According to Dewar & Dutton (1986), innovation is defined as “an idea, practice, or material artifact perceived as new by the relevant unit of adoption” (p. 1422). Yet the degree of innovation to the adopting unit, such as companies, consumer, or markets, varies. Radical innovations, on the one hand, are embodied through technological advances that disrupt an entire market and technology structure to create new structures (Garcia, & Calantone, 2002). They are portrayed as being entirely novel and discontinuous to the current trend of development, unique and unlike any current product, and, maybe most importantly, are accepted by the market and impact further innovations (Norman & Verganti, 2014). If a potentially radical innovation exists but all socio-economic and cultural factors do not align and it is not adopted by the consumers, it cannot be perceived as an actual radical innovation. One example of such a failure of the market to adapt a potentially radical innovation was the introduction of the Microsoft touch input tablet computer (Bort, 2013). Even though the technology was similar, the Microsoft touch input tablet never was adopted by users and did
11
not contribute to further innovations in the same way the Apple IPad did barely one decade later. At the time the innovation might have been too radical for the market. Other famous examples of radical innovation are the World Wide Web, the personal computer, and the smartphone (Norman & Verganti, 2014).
Incremental innovations, on the other hand, are characterized through minor improvements or adjustment to existing technology (Dewar & Dutton, 1986). The major difference between both is described by whether the innovation is perceived as a continuous process, a modification of formerly approved practices, or if it is entirely new, unique, and discontinuous from former processes (Norman & Verganti, 2014).
Hence, when using the description of radical and incremental innovation made above, it becomes clear that the appearance of radical innovation is rather infrequent in comparison to incremental innovation. Especially with regards to the automobile industry, the first electric vehicle arguably has been a radical innovation, may it be the first electric vehicle in the early 1900s or the Nissan Leaf, the first mass-marketed electric vehicle as it is known today (insideevs.com, 2016). Yet in recent years, the advances of the automobile industry have been almost exclusively incremental. Even though new technologies such as hydrogen-powered or autonomous driving vehicles have been introduced or are advancing rapidly, they are not yet accepted by the wider market and can therefore not be accounted for (Luettel, Himmelsbach, & Wuensche, 2012).
Even the most innovative car manufacturer, Tesla, is merely advancing the field rather than disrupting it through radical innovation. However, Tesla and other vehicle producers have significantly influenced the electric and hybrid vehicle sector through their incremental innovations as they are solving some of the major issues facing the industry. For example, as the network of charging stations around the world expands, long-distance travel in an electric vehicle becomes more feasible. Yet in order to facilitate travel over long distances, automobile manufacturers currently compete on extending the range of their vehicles. Furthermore, major advancements have been made in the charging time of electric vehicles batteries and other metrics such as safety scores, sustainability, horse power, acceleration, and overall price of the vehicle (teslamotors.com, 2016). These advancements explain the growth trend in electric and plug-in hybrid vehicle sales in car market around the world.
Based on the information provided above, the mentioning of innovation in the latter sections of this case study therefore have to be viewed with regards to incremental innovation. Furthermore, this research includes electric vehicles as well as plug-in hybrid ones as the field
12
of hybrid vehicles is currently advancing through innovation as types of cars, such as SUV and sports cars, are added to the sector.
As stated earlier, according to Mourdoukoutas & Siomkos (2009), innovation is essential to the creation of a buzz as it sparks consumer interest. For this reason, new introductions of innovative electric or plug-in hybrid vehicles should create a larger buzz than mere trend following cars or extensions of existing lines, modified with a rather non-innovative electric or plug-in version. According to Dye (1999), on the other hand, these vehicles should create a similar level of buzz if marketed correctly. In order to analyze whether innovation is indeed necessary to spark consumer interest and therefore buzz, as stated by Mourdoukoutas & Siomkos (2009), a number of hypotheses are tested. Therefore, as consumer engagement on companies’ social media pages is a proxy for buzz, it is expected that the following hypotheses hold true:
H2: Innovation is positively correlated to consumer engagement during the vehicle launch
H3a: Innovation is positively correlated to the sales during the vehicle launch
Furthermore, as competition in the automobile segment is fierce and innovation is one of the main competitive advantages vehicles manufacturers can gain over their competition, this additional hypothesis is likewise expected to hold true.
H3b: Innovation is positively correlated to the overall vehicle sales
3.3 Visualization of the Conceptual Framework
H1a: Consumer engagement is positively correlated to sales during the vehicle launch
H1b: Consumer engagement is positively correlated to overall vehicle sales
H2: Innovation is positively correlated to consumer engagement during the vehicle launch H3a: Innovation is positively correlated to the sales during the vehicle launch
13
4. Methodology 4.1 Data Collection
The case studies considered in this research are analyzed through a stage multi-method research approach. In the first step a sample is drawn. For this reason, information on new plug-in electric vehicle releases, introduced within the time of the beginning of 2014 until April 2016, is collected. As car manufacturers tend to release new vehicles at different times across various markets, this research will only focus on new electric vehicle releases in the electric vehicle market of the United States of America. As the largest electric vehicle market in the world and due to its large amount of available sales data, it is a reliable proxy for the global electric vehicle market in terms of new product introduction.
A total of six new electric or hybrid vehicle releases by four car manufacturers were identified during the given time frame: The electric BMW i3, Volkswagen e-Golf, Tesla Model X and the hybrid BMW x5 x Drive40e, BMWi8, Audi A3 Sportback e-tron. Additionally, the announcement of the Tesla Model 3 was included into the data even though the vehicle has not been released as this paper is written. Since potential buyers deposited 1.000$ in order to place a reservation, it is probable that the vast majority of said reservations will lead to a sale after the vehicles initial launch. Therefore, the amount of reservations placed for the Tesla Model 3 may act as an estimate for future sales. In the next step, the vehicles are analyzed in order to determine their degree of innovation. For this purpose, qualitative data is collected through the consultation of press releases of the introducing companies regarding the new product introduction, describing the vehicle and its specifications. Additionally, it is analyzed whether the vehicle is part of an entirely new series or solely a continuation of an existing series, a control variable.
Subsequently, data regarding the sales information of the vehicles is collected through the use of the website insideevs.com. The website provides a coherent database of monthly U.S. electric vehicle sales for the given time frame which allows to analyze not only the overall sales of each individual model but also each models performance during its introduction period. With regard to the Tesla Model 3, the web service model3tracker.info, which compiles the latest Tesla announcements from which it drives estimates of the total reservations of the Tesla Model 3 for each country, is consulted.
Finally, information on the U.S.-targeted Twitter accounts of the four analyzed companies was collected for the given time frame through the use of the tool
fanpagekarma.com. Note that while Lassen et al. (2014) analyzed overall consumer behavior
14
each other, Mourdoukoutas & Siomkos (2009) and Dye (1999) consider only the company induced buzz through persuasive messages which spark conversation among consumers, not the resulting overall buzz and conversation. Hence, in order to analyze the ability of a company to create a buzz on Twitter, instead of collecting data on hashtags and consumer messages, data regarding the analyzed vehicle manufacturers’ Twitter account was collected in order to analyze the effectiveness of the company’s messages to create a buzz. In other words, only the ability of vehicle manufacturers to create a buzz through their persuasive Twitter content was analyzed, excluding consumers’, blogs’, or journalists’ tweets containing hashtags regarding the vehicle introduction.
This tool compiles the entire activity and engagement with consumers of the considered company’s Twitter account and rearranges them into weekly reports containing valuable quantitative information such as the overall amount of followers, follower growth, and the total amount of company tweets and consumers’ likes and retweets in the given week. Note that for the purpose of this research only the volume of interactions was used to consider consumer engagement. This procedure is based on the findings of Liu (2006), stating that most of the explanatory power of WOM on sales is derived from the volume of WOM, not the valence. However, the content of tweets that created significantly higher amounts of interactions as well as tweets during the introduction of the new vehicles are examined in order to ensure that the conclusions drawn regarding the source of the buzz are correct.
The secondary data used for this research, the press releases and newspaper articles, the official company sales data available on insideevs.com, and the data based the Twitter accounts through the web tool fanpagekarma.com, is publically and globally available. Due to its degree of availability and source integrity it can be assumed that the quality of the data is reliable (Saunders, Lewis, & Thornhil, 2011).
4.2 Data Analysis
In the first step, the degree of innovation for each individual car is analyzed. Therefore, the collected qualitative data is analyzed for key words such as “world’s first” or “revolutionary” and the overall statistics of the electric vehicle are analyzed in order to determine whether the vehicle can be considered an incremental innovation advancing the entire industry, or simply a regular hybrid or electric vehicle. Furthermore, data is extracted and compiled regarding statistics as well as the price of the car, a control variable used in the further analysis.
Subsequently, the vehicle sales data is analyzed. In order to account for the difference between the vehicles sales performance during the introduction phase and their overall sales
15
performance, first the overall sales performance was analyzed, followed by each vehicle’s individual sales performance during its first year after the launch to the market, the introduction period. In order to analyze the sales performance of the individual vehicles, the monthly average sales overall and the introduction sales are calculated. Moreover, the monthly average sales are controlled for the vehicle list price by calculating the average monthly turnover each vehicle generated. For this research, the lowest list price stated for the individual vehicle by the company was considered, without accounting for government tax credits or subsidies.
Furthermore, it is necessary to mention that the amount of vehicle sales per month was not adjusted to the overall market growth in the electric vehicle market over the past two years. This decision is based on the fact that the United States plug-in electric vehicle market did not perform similarly as the global market. Even though worldwide sales of plug-in electric vehicles have increased by 71.59% between 2014 and 2015 (320.713 vehicles to 550.297 vehicles, respectively), plug-in electric vehicle sales in the United States of America have slightly declined, with total sales of 122.438 vehicles sold in 2014 and 116.099 vehicles sold in 2015 (insideevs.com, 2016; Appendix 1). Yet when considering sales of the current year of 2016, a tendency can be observed that the plug-in electric vehicle market in the U.S. is expected to fluctuate around the baseline set by the previous years, rendering an adjustment to market growth inconsequential. This development may be explained by the fact that the U.S. already are the largest market for electric vehicles and its urban areas, the centers of electric vehicle use, are yet more satisfied than in other nations. Another possible explanation may be the extreme depreciation of oil prices over the past years, decreasing gasoline prices especially in the U.S. and depriving electric vehicles of one of their key advantages in comparison to gasoline powered vehicles.
In the following step, the quantitative data of the companies Twitter accounts weekly performance is analyzed. For this research, retweets and likes, measuring consumer engagement, are considered equally valuable vehicles of e-WOM as both lead to a sharing of the information with the user’s network. Even though retweets are the direct way of presenting the companies information to one’s network, likes occur on the account page of the user, which is publicly visible and equally serves for e-WOM. Furthermore, new factors are calculated to account for control variables such as the Overall Follower Base, Weekly Growth of the Companies Follower Base, and Amount of Tweets per Week. These steps are necessary in order to receive unbiased data. After controlling the data for unexplained outliers, the level of consumer engagement during the individual vehicles announcement and introduction is analyzed. As the data set is non-normally distributed (Appendix 2), a one-sided t-test cannot
16
be conducted in order to determine whether the level of consumer engagement during an introduction significantly deviated from the mean. Bearing in mind the different layers of the data and its integration into a time series with a high likelihood of interdependence between the levels of consumer engagement in the individual weeks, the statistical analysis of the weeks of interest in their position among the total sample population of the individual case studies would exceed the frame of this research paper. Therefore, it was decided to analyze the data solely in a descriptive manner, using the upper bound of the 95% confidence interval as an indicator to determine extreme values. Note that the results of the analysis therefore solely display the tendency of a correlation, not statistical significance.
In order to test for mediation, first the effect of the independent variable, Innovation, on the dependent variable, Vehicle Sales, is analyzed. Therefore, the individual vehicles are clustered into the two segments, Innovation and Trend Follower. The amount of sales and turnover created by each car and the overall category during the vehicles introduction and the overall sales are compared in order to identify tendencies. Note that since to the relatively small sample size of seven electric and plug-in hybrid vehicles the appearance of statistically relevant outcomes is improbable, this research is solely considering the identification of tendencies of correlations.
Ongoing, the interaction of the independent variable, Innovation, and the mediating variable, Consumer Engagement, is investigated. With regards to the clusters formed in earlier steps, the buzz creation of each new vehicle announcement and market introduction is analyzed in order to identify tendencies between the different vehicle models that indicate a significantly larger buzz creation for innovative models than for trend following vehicles. By utilizing the findings of the previous section regarding the interaction of the independent and dependent variable, it is possible to identify a potential mediating effect.
5. Results
5.1 Descriptive Statistics 5.1.1 Innovation
Regarding the degree of innovativeness of the different vehicles, the press releases of the individual companies regarding each vehicle introduction are evaluated and analyzed for key words and statements focusing on the innovations of the vehicle. The results can be found in the section below.
17
The BMW i8 is described as the “world’s most progressive model in the sports car segment” (BMW Group, 2013, p. 1). The BMW i8 is one of the world’s first plug-in hybrid sports cars and continuously described by BMW as “revolutionary”, “future-focused”, and even a “trailblazer for a new generation of sports cars” (2013, p. 6). Apart from performance features such as a superior electric engine which allows a fuel economy below the level of urban subcompacts, BMW highlights the importance of sustainability throughout the entire production chain, with sustainably sourced material and the BMW factory in Leipzig, Germany, powered solely by wind energy. The combination of these features qualifies the BMW i8 as a
truly innovative vehicle. The vehicle was announced on the 10th September 2013 and reached
the U.S market in August 2014.
Similar to the BMW i8, the BMW i3, the second member of the newly established BMW
i sub-brand, focuses on sustainability. According to BMW, the BMW i3 is unlike any previous
vehicle and thorough is “visionary design” and “ground-breaking vehicle architecture” it is meant to call out a new era of electro-mobility (2013, p. 1). The vehicle is powered solely by electricity and composed of the “first-ever mass produced carbon fiber reinforced plastic (CFRP) passenger cell”, which is as strong as steel, yet 50% lighter. Furthermore, 25% of the interior and exterior plastics are extracted from recycled materials or renewable sources, and, apart from being produced in the previously mentioned energy efficient factory in Leipzig, Germany, the CFRP components are sustainably produced in Moses Lake, WA, USA, a factory entirely operated using hydroelectric power (BMW, 2013). As the BMW i8, the i3 also
qualifies as innovative vehicle. The vehicle was announced on the 29th July 2013 and reached
the U.S market in May 2014.
On the other hand, the BMW x5 eDrive40, does not qualify for the innovative category. Even though it is the “first plug-in hybrid BMW sports activity vehicle”, rather than an innovation, the BMW x5 eDrive40 is the sole extension of the classic BMW x5 series including
a hybrid option (BMW, 2015). The vehicle was announced on the 16th March 2015 and reached
the U.S market in October 2015.
The same accounts for the VW e-Golf. At a starting price of 33.450$ the vehicle may be the most cost-effective model analyzed in this research, yet rather than fostering innovation, the VW e-Golf counts to what BMW describes as “conversion vehicles”, the replacement of a gasoline powered engine in a previously existing vehicle with an electric engine (BMW, 2015).
Announced on the 25th August 2014 and launched to the U.S. market in October 2014, the
company claims that the vehicle represents a “groundbreaking and holistic approach to e-mobility”. Yet rather than investing in sustainable production, Volkswagen simply invests in
18
other carbon reduction projects outside the company in order to offset the emission created from the production (Volkswagen, 2015). Even though this production approach is commendable and environmentally friendly, it accomplishes no technological advancement regarding the sustainability of the production itself. Therefore, the vehicle cannot be considered innovative.
Similar to the VW e-Golf and BMW x5 eDrive 40, the Audi A3 e-tron displays no sign of innovation and is simply the extension of the classic Audi A3 series featuring an electric
engine. Announced on the 5th August 2015 as “latest generation plug-in hybrid car” and
launched to the U.S. market in December 2015, the Audi A3 e-tron is counted to the trend following category.
The Tesla Model X, on the other hand, can be seen as a truly innovative car. Announced
on the 9th February 2012 and launched to the U.S. market in September 2015, Tesla claims that
the Model X is the “safest car on the road, receiving five star safety ratings in every category, the first SUV ever to do so” (2016, p. 1). Furthermore, the all-electric Tesla Model X displays an extraordinary driving range of up to 414km on a single charge and is the “first electric vehicle with a 5.000 lb. tow capacity, the strongest in the industry” (Tesla Motors, 2016, p. 1).
Ultimately, the Tesla Model 3, announced on the 31st March 2016, includes key features
of other Tesla models, such as a wide range of up to 320km with one charge and the claim of being the safest vehicle of its category. However, through a superior and innovative production process Tesla is able to distribute this premium vehicle at a starting price of a mere 35.000$, rendering it the most cost-efficient vehicle in the electric vehicle market (Tesla Motors, 2016).
5.1.2 Product Sales
In order to investigate the effect of the independent variable, innovation, discussed in the section above, on the dependent variable, the vehicles sales during the introduction as well as overall sales, the data extracted regarding the monthly sales of the newly introduced vehicles is consulted. A summary of the prepared data can be found in Table 1.
The data is ordered by degree of innovation, with the three innovative models in the top, and the three trend following models in the bottom half. Furthermore, the product sales were divided into two time periods, one including only the time period of the first year of the new products introduction, the introduction phase, while the latter table includes all sales since the individual vehicle’s introduction. This division is necessary in order to control for differences between overall sales and sales during a vehicles introduction where buzz could be
19
especially influencing. Furthermore, sales are analyzed by the average amount of product sales per month, as the vehicles were introduced at different moments within the time frame.
Regarding the introduction period, the BMW i3, the vehicle with the lowest price in the innovative category, sales by far the most models with a mean of 765 vehicles per month, amounting to an average monthly turnover of 31.629.304$. The BMW i8, the sports car of the BMW-I-series, scores average monthly sales of 125 vehicles.
Table 1: Product Sales during the Introduction and Overall; by Model
Yet as the most expensive vehicle in the category, 135.700$ per vehicle, the mean monthly turnover results to 17.019.042$. The highest turnover in the category of innovative vehicles is achieved by the Tesla Model X, with average monthly sales of 433 vehicles at a price of 132.000$, resulting in an average monthly turnover of 57.156.000$ during the period of its introduction.
The Audi A3 sports back e-tron, a trend following vehicle, sold on average 256 units per month in the United States. With a list price of 37.900$ per vehicle, these sales accumulated to an average monthly turnover of 9.717.560$ during its introduction. The BMW X5 x Drive40e, a plug-in hybrid extension of the successful BMW x5 series, sold on average 289 vehicles per month at a price of 62.100$ per vehicle, accounting for a mean monthly turnover of 17.915.850$. The VW e-golf sold on average 291 units at a price of mere 33.450$ during its introduction, scoring an average monthly turnover of 9.730.429$.
20
The Tesla Model 3, a vehicle that although it has only been announced but not released to the market yet, has already been reserved more than 400.000 times across the globe, with approximately 223.319 reservations made only in the United States (model3tracker.info, 2016). At a starting price of 35.000$ and assuming that the 1.000$ deposit required to reserve the vehicle provides sufficient incentive for consumers to actually purchase the car once it is launched, the total turnover only in the U.S. market would accumulate to 7.816.165.000$. This figure will be utilized in the later analysis as stating solely the 1.000$ deposit would not accurately reflect the sales potential of the vehicle and its main innovation, its cost-effectiveness. However, this data is based on the assumption that a reservation consequently leads to a sale. Furthermore, as no information is available regarding Tesla’s production capabilities with respect to the Model 3, no estimate can be made on the monthly sales once the vehicle is launched to the U.S. market and therefore it cannot be effectively compared to the other vehicles in this study. Nevertheless, the estimated total turnover serves as an indicator for the anticipated overall sales of the Tesla Model 3 during its introduction and is therefore used in this research.
Note that the Tesla Model X and Audi A3 e-tron are still within their introduction period, hence its sales information regarding the two time periods is identical. Furthermore, a possible factor influencing the fluctuations in electric and plug-in hybrid vehicle sales could be legislative changes and government incentives subsiding sales of electric vehicles, a variable not accounted for in this research.
5.1.3 Consumer Engagement
In order to analyze the consumer engagement with company injected marketing messages regarding announcement and introduction of new electric and plug-in hybrid vehicles, the data collected on the Twitter activity of the company accounts of Audi, Tesla, BMW, and Volkswagen is consulted. In order to analyze the data properly, first the overall follower base is observed as well as the growth in followers during the period of investigation. Subsequently, the average consumer engagement of the companies is investigated in order to determine the mean levels of engagement and presented alongside other relevant descriptive statistics.
In the graph considering the overall follower base of the individual companies and its growth over time (Appendix 3) it can be observed that the amount of overall followers of the different companies accounts appear to be relatively similar with exception of the company Audi. As Coulter & Roggeven (2012) stated in the literature review above, a large amount of pre-enrolled members of a network signal more credibility which, in return, affects consumer
21
knowledge and liking regarding the company. However, in the case of Audi it is possible that the greater follower base is derived from the fact that the Audi’s Twitter account targeting the United States market is simply named @audi, a name users from countries which Audi is not targeting with individual account are most likely to follow. The same accounts for Tesla Motors, as it is a relatively young company and therefore channeling its buzz creation power towards only one Twitter profile.
Another control variable, demonstrated in the Appendix 4, the relative growth of the respective companies’ follower base, needs to be accounted for in order to prevent a scrutiny of outcomes towards the later dates of the analyzed time period. Hence, in order to account for the control variables “overall follower base” and “relative growth of follower base”, the total amount of interactions per company are divided by the amount of followers each week and multiplied by 100.000, in order to achieve the standardized output “Interactions per 100.000 Followers”. The difference between solely adjusting for growth and adjusting for the follower base and growth is displayed in the Appendix 5.
Note that in this research, the average amount of Tweets tweeted by the company account was not regarded as a control variable. Controlling for the average amount of interactions per tweet would dilute the results of this study, as the nature of these tweets is often uncertain. For example, it is possible that during the launch week the company responded to consumer questions regarding the product through tweets, which in exchange only received little interactions by other consumers unaffected by this inquiry. As these tweets would distort the data significantly this control variable was not accounted for in the analysis. However, from the table included in the Appendix 6, it can be observed that while BMW and Tesla Motors are equally active on Twitter (average of 25 and 27 Tweets per week, respectively), Audi (average of 55 Tweets per week) and Volkswagen (average of 78 Tweets per week) are significantly more active senders of Twitter messages than the former two vehicle manufacturers. This difference will be addressed later in the discussion of results section.
Regarding the descriptive statistics, Volkswagen’s Twitter account @vwusa incurs a mean of 506 interactions per week with a 95% confidence interval with an upper bound of 625 weekly interactions (Table 2.1). It is expected that weeks containing specific buzz-generating events will display a level of interactions above this upper bound of the 95% confidence interval, an indicator for their extreme deviation from the mean. The Twitter account
@teslamotors generates an average of 1094 interactions per week with its upper bound of the
22
@audi displays a mean of 629 weekly interactions with an upper bound of the 95% confidence
interval at 688 interactions per week (Table 2.2).
In the case of BMW, a mean of 1314 interactions per week can be observed. However, the sample range included weeks with a maximum of 22012 interactions (Table 2.2). These outliers, portrayed in the Appendix 7, can be partially explained by BMW’s sponsorship of the United States Olympic Bobsled team in early 2014, yet its extent is unprecedented. Consequently, it was decided to remove the three outliers from the @bmwusa data set due to the distorting nature of these values. The newly calculated mean for the adjusted @bmwusa data set lies at 961 interactions per week with the upper bound of the 95% confidence interval at 1116 weekly interactions (Table 2.3).
Table 2.1: Descriptive Statistics of the Volkswagen’s and Tesla Motor’s Twitter Accounts
23
Table 2.3: Descriptive Statistics of the BMW Twitter Account, adjusted for outliers
5.2 Hypotheses Testing
5.2.1 Correlation between Innovation and Vehicle Sales
For the purpose of analyzing the correlation between innovation and vehicles sales, the product sales data regarding the vehicles sales is grouped by category and presented (Table 3.1).
Table 3.1: Product sales during the Introduction and Overall; by Category; excl. Tesla Model 3
During the introduction phase, innovative vehicles sold a mean amount of 441 vehicles at an average price of 103.017$ per vehicle, accumulating into an average monthly turnover of 35.268.115$ per month. Trend following vehicles, on the other hand, sold on average 263 vehicles per month at an average price of 44.483$ per vehicle. Their sales lead to an average turnover of 11.916.870$ during the introduction period.
Considering the overall sales, the innovative vehicles sold on average 453 units per month, accumulating to an average monthly turnover of 36.483.612$. The trend following vehicle category scored mean monthly sales of 296 vehicles, leading to a mean turnover of 13.538.406$ per month.
24
First, it can be observed that innovation appears to come at a price. Vehicles of the innovative category sell at an average premium of 58.534$ in comparison to the trend following category. Yet during the introduction period innovative vehicles sold on average 178 vehicles or 68% more per month than trend following vehicles. In terms of overall sales, innovative vehicles still sold 157 units or 53% more per month than the trend following category. Expressed in average monthly turnover, vehicles of the innovative category generated 23.351.245$ more in turnover, roughly 196% more than the trend following category, during the introduction period. This gap normalizes slightly when analyzing the overall product sales. However, innovative vehicles continued to generate 22.945.206$ more in average monthly turnover, which accounts for 169% more turnover than the trend following category. Hence, it appears that consumers are willing to pay the above mentioned premium for vehicles of the innovative category, whether it be due to the brand recognition, vehicle features, or perhaps the innovativeness of the vehicles.
Furthermore, when analyzing the individual vehicle sales stated in previous section (Table 1), another tendency emerges. When comparing the individual vehicle sales during the introduction period to the overall sales, marginal increases in sales of the BMW i3 and i8 can be observed, suggesting a slightly higher degree of market acceptance after its introduction period. Yet the BMW X5 xDrive40e and the VW e-golf, vehicles of the trend following category, increased their overall sales by 18% and 20% in comparison to the introduction period. This could be seen as an indicator that trend following cars require longer time periods in order to raise consumer awareness and foster sales in comparison to innovative vehicles.
The superior sales potential of innovative vehicles over trend following vehicles becomes especially clear when including the Tesla Model 3 into the analysis (Table 3.2). Even though the results of this comparison are questionable until the official market launch of the Tesla Model 3, expected in 2017, the ability of the vehicle to revolutionize the electric vehicle market is evident and reflected in its sales potential.
It appears that vehicles of the innovative category tend to create significantly higher sales per unit and turnovers, especially during the introduction period. Hence, it can be concluded that hypothesis 3a is supported, as a tendency of innovative vehicles to generate higher vehicle sales during the introduction period can be observed. Furthermore, innovative vehicles are observed to generate a significantly higher amount of sales overall, even though to a marginally lower extend, supporting hypothesis 3b.
25
Table 3.2: Product sales during Vehicle Introduction and Overall; by Category; incl. Tesla Model 3
5.2.2 Correlation between Innovation and Consumer Engagement
The following section focuses on hypothesis 2, a possible positive correlation between innovation and consumer engagement. For this reason, the amount of consumer interactions the individual companies accounts incurred on Twitter are visualized below in the form of graphs. The x-axis of the graphs represents the individual weeks expressed as continuous calendar weeks during the sample time frame (January 2014 until April 2016). The y-axis represents the amount of consumer interactions, likes and retweets, per 100.000 followers the respective company incurred in any given week.
As observed in the descriptive statistics, Volkswagen’s Twitter account @vwusa engages on average in 506 interactions per week, with the upper bound of the 95% confidence interval lying at 625 interaction per week. During the announcement week of the VW e-Golf only 388 interactions were accounted for, an amount on the line of the lower bound of the 95% confidence interval (Graph 1). During the market launch 575 interactions were counted, an amount within the confidence interval. The VW e-Golf therefore failed to create a significant buzz surrounding its introduction. As the vehicle is counted to the trend following category, these findings were expected under hypothesis 2.
26
Graph 1. Weekly amount of consumer interactions generated by Volkswagen’s Twitter account @vwusa
Tesla’s Twitter account @teslamotors counts an average of 1094 interactions per week with the upper bound of the 95% confidence interval at 1274 interactions per week. During the week of the launch of Tesla’s model X, simply announced as “Meet model X!” on twitter, 1916 interactions where counted, creating significantly higher levels of consumer engagement than on average (Graph 2).
Graph 2. Weekly amount of consumer interactions generated by Tesla Motor’s Twitter account @teslamotors
During the week of the announcement of the Tesla Model 3 on the 31st March 2016, a
total of 6349 interactions were counted, largely exceeding the boundaries of the confidence interval. Bearing in mind the innovative features of these vehicles and their contribution to the
27
field of electric vehicles, the higher levels of consumer engagement support the assumptions made in hypothesis 2.
BMW’s Twitter account @bmwusa, adjusted for three outliers, on averages counts 961 interactions per week with an upper bound of the 95% confidence interval at 1116 interactions per week. During the launch of the BMW i3 and i8, vehicles of the innovative category, up to 3019 and 1951 interactions were counted, respectively (Graph 3). During the announcement of the BMW X5 x eDrive40, a trend following vehicle, only 394 interactions were counted, during its first launch week a mere 402 interactions. Given these observations, both innovative vehicles scored significantly higher in their levels of consumer engagement than the trend following vehicle, as it is expected by hypothesis 2.
Graph 3. Weekly amount of consumer interactions generated by BMW’s Twitter account @bmwusa; adjusted
As described in the section above, Audi’s weekly mean interactions on its Twitter account @audi are 629 with an upper bound of the 95% confidence interval at 688 interactions
per week. The week of the announcement of the Audi A3 e-tron 858interactions are counted,
170 interactions above the boundaries of the 95% confidence interval (Graph 4). However, the vehicle was announced solely one week prior to the announcement of the Audi R8, Audi’s flagship model, which may have considerably influenced the level of consumer engagement during the A3 e-tron’s introduction. The first launch week, when Audi announced the vehicle
28
stating “Introducing the plug-in hybrid you won't be ashamed driving to Las Vegas—the #AudiA3 #etron”, a total 715 interactions can be counted, slightly above the confidence interval.
Graph 4. Weekly amount of consumer interactions generated by Audi’s Twitter account @audi
These findings stand opposing to hypothesis 2 regarding influence of innovation on consumer engagement during new vehicle introductions as the Audi A3 e-tron belongs to the trend following group, yet creates a higher level of consumer engagement than the mean. The increased level of consumer engagement may be explained through advanced marketing efforts of Audi regarding the A3 e-tron, or due to the fact that during the announcement one of Audi’s flagship models, the new Audi R8, was introduced almost simultaneously, increasing overall traffic on the @audi Twitter page. However, in comparison to the levels of consumer engagement of innovative vehicle introductions, the announcement of the Audi A3 e-tron (858 interactions) and its launch to the U.S. market (715 interactions) still created observably lower levels of consumer engagement than the launch of innovative cars such as the Tesla Model X (1916 interactions), the Tesla Model 3 (6349 interactions), the BMW i3 (3019 interactions), and the BMW i8 (1951 interactions).
The table below summarizes the results of the analysis conducted above (Table 4). The
data collected on the Twitter accounts portrays a clear picture. The four innovative vehicles considered in this research, the upper half of the table, consistently created larger amounts of consumer engagement during their market launch than the account average and exceeded the upper bound of their respective 95% confidence interval.
The vehicles of the trend following category, displayed in the lower half of the table, generated significantly lower levels of interactions during their market launches and only the
29
Audi A3 e-tron generated an above average level of interactions and consumer engagement. Yet, as mentioned above, the level of consumer engagement created by the Audi A3 e-tron remains considerably lower than the level of consumer engagement generated by the vehicles of the innovative category. Furthermore, it is unclear to what extent the introduction of the new Audi R8, Audi’s flagship sports car, has influenced the level of consumer engagement in the previous week during the announcement of the Audi A3.
Table 4: Comparison of mean interactions with interactions during vehicle announcement and market launch
Therefore, with respect to the findings mentioned above, a tendency of innovative vehicles to create higher levels of consumer engagement has been identified, supporting hypothesis 2.
5.2.3 Correlation between Consumer Engagement and Vehicle Sales
In this segment, the correlation between consumer engagement and vehicle sales is analyzed. Since the number of interactions is adjusted to a standard of “per 100.000 Followers”, it is possible to compare the level of average consumer engagement of each vehicle introduction to the average turnover made during the introductions, first for the individual case studies and then the two separate categories. In order to examine the perceived correlation between the moderating variable Consumer Engagement and the dependent variable Product Sales the data regarding the two variables was summarized in Table 5.1.
30
The BMW i3 generated the largest amount of interactions, 3019, during its introduction week, with 31.629.304$ in average turnover per week, followed by the BMW i8 with an amount of 1951 interactions during its introduction week and 17.019.042$ (Table 5.1). The Tesla Model X generated the third largest amount of interactions during its introduction week, 1916, but the largest average monthly turnover, 57.156.000$. The vehicles with lower levels of interactions during their introduction, the Audi A3 e-tron (715 interactions) and VW e-golf (575 interactions), generated a lower average turnover of mere 9.717.560$ and 8.117.200$, respectively. Yet the BMW X5 x Drive, which generated only 402 interactions during its introduction, the lowest amount of interactions of all seven case studies, sold more vehicles than the two previously mentioned models, accumulating to average monthly turnovers of 17.915.850$. Hence, apart from the BMW X5 x Drive, a tendency regarding the correlation between consumer engagement during a vehicles launch and the average monthly turnover during its introduction can be observed. When analyzing the overall sales of the individual vehicles over the entire time frame of this study, this trend stabilizes (Appendix 8).
However, in order to explain the counterintuitive mismatch between low amount of interactions generated during the introduction of the BMW X5 x Drive and its relatively higher sales, another correlation is hinted. When observing the average consumer engagement created by the individual companies Twitter accounts throughout this study, it can be seen that the producers of innovative cars, BMW and Tesla Motors, generated higher average levels of weekly interactions (961 and 1094 interactions, respectively) and therefore consumer engagement than the manufacturers of the trend following vehicles, Volkswagen (506 interactions) and Audi (629 interactions). These findings suggest a tendency for a correlation of high levels of overall consumer engagement of companies on Twitter with higher levels of sales.
When considering the vehicles’ respective categories, the correlation becomes clearer. As previously mentioned, vehicles of the innovative category generated 35.286.155$ in turnover, roughly 196% more than the trend following category, during the introduction period. However, as it can be seen in table 5.2, innovative vehicles furthermore generated on average 2.295 interactions during their individual introduction, roughly 307% more than the vehicles of the trend following category. Furthermore, in the average overall consumer engagement generated by the companies, the innovative vehicle manufacturers displayed an average level of consumer interactions of 1.005, 23% higher than the companies producing trend following vehicles. Even though the gap between the vehicle sales aggregate slightly when analyzing overall sales (Table 5.2), the difference is constant.
31
Table 5.2: Comparison of vehicle sales and consumer engagement during introduction and overall; by category
When including the Tesla Model 3, the tendency becomes more significant. As displayed in (Table 5.3), reservations of the Tesla Model 3 during its introduction have accumulated to a total estimated turnover of 7.816.165$. Apart from generating the highest turnover of all vehicles introduced in the past two years, Tesla Motors furthermore incurred 6349 interactions during the announcement of the Tesla Model 3, almost five times higher than its average level of consumer engagement. When analyzed with regard to its belonging to the innovative category (Table 5.4; Table 5.5), it can be shown that the Tesla Model 3 significantly increases the average amount of total sales and therefore average turnover. Furthermore, by adding the Tesla Model 3 into the analysis, the amount of average interactions during the average innovative vehicle’s launch increases 44%, further supporting the tendency observed above.
Considering the evidence derived from these case studies, a clear tendency can be seen regarding the relationship between consumer engagement and vehicle sales. Vehicles with higher levels of consumer engagement during their introduction are therefore believed to generate higher average turnovers during their introduction and overall, supporting Hypothesis 1a and 1b. Furthermore, the data, especially the sales of the BMW X5 x Drive, displayed a tendency of a positive correlation between the overall consumer engagement created by a company and their vehicle sales, regardless of the degree of innovation the vehicle represents.
32
Table 5.4: Vehicle sales and consumer engagement during introduction; by category; incl. Tesla Model 3
Table 5.5: Overall vehicle sales and consumer engagement; by category; incl. Tesla Model 3
5.2.4 Mediating Effect of Consumer Engagement on Innovation and Vehicle Sales
Considering the above stated results regarding the different hypotheses, a tendency of the different variables influencing another can be observed. The analysis displays the effect of the independent variable Innovation on both the dependent variable, Vehicle Sales, and the moderating variable, Consumer Engagement. Both variables are positively correlated, demonstrating the positive effect of Innovation on both buzz creation and average vehicle sales during a vehicle’s introduction. Yet, a vehicle’s buzz creation ability during its introduction cannot entirely explain vehicle sales, as it is displayed by the low level of engagement during the BMW x5 x Drive40e’s introduction and its relatively high average monthly turnover. This observation could be explained by a possible correlation between a company’s mean level of consumer engagement and overall vehicle sales.
Since the overall level of buzz surrounding a company on social media cannot be attributed to a single event but rather to a mix of continuous innovation, vehicle releases, and marketing efforts, the exact degree to which the individual vehicle’s innovation influences the overall level of consumer engagement remains unclear. However, from the results and tendencies presented above, it can be concluded that the mediating variable Consumer
