First movers and fast followers complementary resource and market differences : a double case-study in the computer industry

Hele tekst

(1) . MSc thesis strategy track . First movers and fast followers complementary resource and market differences . A double case‐study in the computer industry Student: Bastiaan Koster 10409378. Supervisor: Ranjita Singh. Institution: University of Amsterdam, Faculty of Economics and Business Date: July 2014 . .

(2) Contents Contents . 2 . Abstract . 3 . 1. Introduction . 4 . 2. Literature review and research questions . 6 . 2.1 First and late mover advantages . 7 . 2.2 First and later movers competence and resource differences . 10 . 2.3 First and late mover advantages in developed and emerging markets . 13 . 3. Methodology . 18 . 3.1 Research design . 18 . 3.2 Case selection . 19 . 3.3 Selected Cases . 21 . 3.3.1 RQ1 Selected cases . 21 . 3.3.2 RQ2 Selected Cases . 23 . 3.4 Data sources . 25 . 3.5 Data Analysis . 26 . 4. Results . 30 . 4.1 Summary of results for RQ1 . 30 . 4.2 Summary of results for RQ2 . 42 . 5. Discussion . 51 . 6. Limitations and future research . 54 . 7. Conclusions . 55 . References . 56 . Appendix: Coding scheme . 58 . . Page 2 of 62 .

(3) . Abstract First mover and late mover advantages are an important topic that has received considerable research attention. I have spotted two areas within this field that have not gotten a lot of attention yet. Relatively little research has been done linking first mover (dis)advantage theory with the resource based view. We know too little about what resources are needed to create a successful first mover or fast follower product. A second lack in research is on the impact emerging markets have on first mover and late mover advantages. While these markets are becoming more important particularly in the automobile and smart‐phone industry, where China is now the biggest market. In this thesis I have done a separate case study for each of these two subjects. My first case study shows that first movers and fast follower products were created using different internal and external complementary resources. When creating first mover products the companies I studied have outsourced critical manufacturing resources. They focused instead on owning a few critical complementary resources related to the product and the consumer. Fast follower products were created with complementary manufacturing resources that the fast‐ follower already owned. The fast‐follower companies were also more diversified companies. My second case study showed that market differences between a developed and emerging market have a big impact on first mover and late mover advantages. The impact of the factors in two theoretical models were studied.   . Page 3 of 62 .

(4) 1. Introduction First mover advantage is an ambiguous concept but it is generally seen as the advantage a company has when entering a market first (Lieberman and Montgomery, 1988). Initially the research findings were mixed regarding the existence of first mover advantages. Some research found its existence while other research could not find first mover advantages. The concept of first mover advantages turned out to be more complex than initially thought. In Lieberman and Montgomery (1988) the concept of late mover advantages was presented as moderator for first mover advantages. While first movers had advantages late movers had them also. Kerin et al (1992) went further on this concept by introducing a framework that showed that first mover advantages and late mover advantages are moderated by many environmental factors. Lieberman and Montgomery (1988, 1998) in their papers asked for future research to focus on important gaps in their theory. First of all there was the call to link their theories with the concepts of the research based view. Being a first mover or a follower could be related to the resources a company has. A second call was for more attention to be paid to the impact of emerging markets. Because markets differ this could impact the fit that pioneers or followers have with it. As I will show in the literature section that some research has already been done on both of these two topics. But a comprehensive knowledge on these subjects is still lacking. That is why I address both these subjects in this thesis. I address each of these two topics with a research question with related propositions and a separate case study to answer each of these questions.  The following chapter of this thesis will start with a literary review where I will introduce my research questions based on the gaps in knowledge I have found. After this I will explain the qualitative case study methods I have used to come to my conclusions. In the results section I will describe the findings of each case and answer the propositions that I introduced in literary review section. In the discussion section I . Page 4 of 62 .

(5) will relate my findings to other research. I will end the thesis with a conclusion. Besides the scientific relevance my thesis will be valuable to both managers and investors when evaluating possible first‐mover (dis)advantages.   . Page 5 of 62 .

(6) 2. Literature review and research questions In this chapter I will first introduce two very important papers that form the basis of the first mover advantage literature by Lieberman and Montgomery (1988) and Kerin et al (1992). These papers show the advantages that first movers and later movers can have. I will show that the current literature lacks research on two topics. First little research has been done on the resources and capabilities that first movers and followers need to have. Second there has also been little research on the impact of emerging markets on first mover and follower advantages. One important paper about each of these topics has already been written by Robinson et al (1992) and Nakata and Sivakumar (1997) but I will show that they only partly fill the gap in knowledge. The first paper links the first mover advantage literature to the RBV and the second predicts the impact of emerging markets on first movers. The aim of my research is to further fill the gap in knowledge that still exists.  I will end this chapter with my research questions and propositions that are based on the gap in literature.. Page 6 of 62 .

(7) 2.1 First and late mover advantages Much of current research on first mover advantages builds on the article that was written by Lieberman and Montgomery (1988). They proposed that pioneering firms had certain advantages because of technology leadership, preemption of assets and buyer switching costs. Technology leadership comes from having proprietary learning opportunities and winning patent races according to Lieberman et al (1988). Preemption of assets can be about input factors or   geographic location and product space. Lastly a pioneer can create buyer switching costs. These costs can be related to transaction costs, customer based learning and contractual agreements. This means that a customer will have to invest more time and/or money to buy from a new seller.   According to Lieberman and Montgomery (1988) late movers also have certain advantages. Firstly there can be a free‐rider effect where late movers can get an advantage by what the first mover have already created. An example of this is the use of a distribution network created for another company. Second of all a first mover takes on risk and uncertainty which can be resolved for a late mover. Lieberman et al (1988) give the example of the emergence of a dominant design. Thirdly a follower firm can take advantage of changes in technology and consumers preferences. Lastly a first mover can become inert allowing later movers entry.   Kerin et al (1992) used the Lieberman et al (1988) paper as a basis to create a more extensive framework (see figure 1). This framework shows that first‐mover (dis)advantage is complex subject to many moderating factors. In this framework a first mover advantage is caused by a cost and/or differentiation advantage that is caused by four factors. A first mover advantage can be economic, an example is scale economies but this is moderated by various market forces such as demand uncertainty, entry scale, efficient scale to market size, advertising intensity, response time and scope economies. A first mover advantage can be preemptive e.g eliminating competition by making a first move, but market and product characteristics moderate this effect. A first mover . Page 7 of 62 .

(8) advantage can also be technological but this is moderated by technological innovation characteristics and technological change and discontinuity of technology. Lastly a first mover advantage can be based on behavioral factors, which means the  possibility to create product differentiation. This is also dependent on many factors, the nature of the product, nature of the market, evolution of the market and buyer investments in co‐specialized assets. According to the Kerin et al (1992) model companies can also have late mover advantages. It is often cheaper to imitate than to innovate. A later mover can take advantage of free‐rider effects if it can profit what the first mover created. A late mover can take advantage of scope economies if the imitated product is compatible with the current operations of the follower. Lastly a late mover can learn from the mistakes that first movers make. In this model first mover advantage and late mover advantages are related, when there are stronger late mover advantages the first mover advantage is smaller because the pioneer and late mover compete with each other. . Page 8 of 62 .

(9) Figure 1: Framework for First mover and later mover advantages by Kerin, Varadarian and Peterson (1992 page 39). . Page 9 of 62 .

(10) 2.2 First and later movers competence and resource differences In the RBV companies compete with each other to obtain costly to copy resources. According to Barney (1991) these resources should hold VRIN characteristics (valuable, rare, inimitable and not substitutable) according to Peteraf (1993) the resources should have limits to ex ante and ex post competition and be inimitable and immobile. Teece (1986) showed that to commercialize an innovation complementary resources are needed. This finding is important to both first mover and later movers because both are bringing an innovation to the market. Complementary resources are resources that are needed together to profit from an innovation like production capability and brands. Relatively little attention has been given to combine the theories of first mover (dis)advantage with the resource base view. Lieberman and Montgomery (1988, 1998) think that more research should be done to combine the fields of first mover (dis)advantages and the resource based view. Because the first mover advantage literature should not risk being vague and definitionally elusive (Lieberman and Montgomery, 1988) by lacking rigor and connection with other theories in strategy. Until now the resource based view and first mover (dis)advantage have been developed independently. The theories seem compatible, according to Lieberman and Montgomery (1998) there are strong synergies to combining the theories. Knowledge from the first mover (dis)advantage literature and the RBV can enhance and strengthen each other. Currently there is a lack of micro‐foundations telling us what resources and capabilities first movers and later companies have. We need to combine first mover and later mover advantages with the RBV to know if these companies have similar or dissimilar resources. It could be that a typical resource set is needed for an innovator while a follower could need a different one. It could be that some companies are better suited to pioneer while other companies have the right resources to follow others. . Page 10 of 62 .

(11) An exception to the lack of research on pioneer and follower resources and capabilities is the research by Robinson et al (1992). They showed that pioneers are not just stronger companies (absolute pioneers advantage hypothesis) and that follower companies differ in resources from first movers (comparative advantage hypothesis). To measure strength they asked managers to self report their functional skills and looked at financial and non‐financial resources that companies had. They classified companies as pioneers, fast followers and late movers. First movers tended to be less well known brands with no link to a parent brand, because pioneering is risky. Established companies do not want to risk their image on a new product that can fail. Fast followers tended to be companies with good manufacturing capabilities that are related to the product they imitate. Late movers tended to be small companies with good marketing resources to occupy a niche. The research was done with quantitative methods using a PIMS database using 177 diversification efforts. PIMS stands for profit impact of market strategy. The unit of analysis of this database is the strategic business unit (pimsonline.com). My thesis will focus on first movers and fast followers because they have been very successful business‐models in the computer industry, my cases showing this are presented in the methodology section.   The research by Robinson et al (1992) gives some insight into the resources and capabilities of companies that enter a market first and those that follow the leader. But there are still gaps in this knowledge. Robinson et al (1992) showed that fast followers had a complementary manufacturing capability but he did not research how first movers produced their products. Teece (1986) showed that to commercialize an innovation complementary resources are needed. Complementary resources are resources that are needed to profit from an innovation like production capability. But these resources can be internalized or bought from other companies. A first mover firm will face high cost and high risk (Robinson et al, 1992). It is much more expensive to create a new product than it is to copy it. A new product is also risky because it can fail, the first mover can not be sure . Page 11 of 62 .

(12) that the customer will like the product. This is why a first mover is likely to buy complementary resources from other companies. The costs of developing complementary assets are an important determinant of the focus of the firm’s business model according to Jacobides et al (2006), outsourcing can also lower the risk because a company does not need to invest in its own resources. Outsourcing makes it possible for a pioneer to focus on its core capability which is disrupting the current market with an innovative product. Robinson at al (1992) have already shown that a fast follower will have a production capability that is related to the pioneer product. In this way they can use synergies and create products at low cost. Secrecy and lead time are important ways for an innovator to protect its innovation (Cohen et al, 2000), it can be expected that fast followers can use their production capability to gain valuable knowledge about a first mover innovation via spillovers, especially if the first mover and follower cooperate with each other. A fast follower does not have the same costs and risks as a first mover (Robinson et al, 1992), because its cheaper to imitate and the business success of the product can already be proven (Kerin et al, 1992). Fast followers will be more diversified companies because they do not know in advance which pioneer will be successful and what technology to follow. Their production capability can be used to create a broad range of products to leverage their resources.       RQ1: How do first movers and fast followers differ in the resources and capabilities they need to commercialize their innovation?   Proposition 1:  In the computer industry first movers will focus on product innovation and outsource complementary manufacturing resources, fast followers will own a diversified complementary production capability . Page 12 of 62 .

(13) 2.3 First and late mover advantages in developed and emerging markets A second gap in the first and later mover advantage literature relates to differences in markets. The empirical evidence regarding first mover advantages largely comes from the United States. Lieberman and Montgomery (1998) believe that more research should be done regarding international markets because research suggest there are differences between countries. Calantone, Song and Di Benedetto (2010) show that the perception of first mover advantages differ between countries and is caused by different mental models impacted by culture and local competitive situations. According to Alpert et al (1996) more than half of the new offerings to supermarkets come from pioneering brands in Japan compared to only 14% in the USA.   Especially the differences between developed markets and emerging markets could be big because of the large economic and cultural differences. Emerging markets are an important phenomenon for companies because of the large population with interesting demographics in combination with high growth rates. The growing middle class in emerging markets means that many people are able to consume more products (Nakata and Sivakumar, 1997). China is already the biggest market for cars and since 2013 also for smart‐phones, both positions were taken over from the United States with a richer but much smaller population.   Little research has been done on the impact of emerging markets on first mover advantages one exception is a paper by Nakata and Sivakumar (1997). According to the authors an emerging market is characterized by two opposite factors, there is a lack of development but there is also a chance at progress in this development coupled with high growth. This means emerging markets are at the same time a big risk and big opportunity for first movers. They used the framework of Kerin et al (1992) and proposed that there were both advantages and disadvantages in emerging markets for all the elements of the Kerin et al (1992) framework, which were: economic factors, . Page 13 of 62 .

(14) preemption factors, technological factors and behavioral factors. Nakata and Sivakumar (1992) propose that first mover and late mover advantages can be impacted both positively and negatively by the circumstances in emerging markets based on economic, technological, socio‐ cultural, legal/political, and competitive/marketing conditions (see figure 2).  A first mover can profit from economic factors in an emerging market because growth is high, a dual economy creates a rich elite and there is a growing middle class consumer. But there are also economic factors that hinder first movers like high inflation and high external debt. A first mover will have to deal with negative technological factors because the infrastructure can be underdeveloped and distribution channels fragmented. But a first mover can profit from these technological circumstances by building its own proprietary distribution systems. A first mover can be negatively influenced by social cultural factors in emerging markets, there is large cultural fragmentation creating many market segments and its hard to find the right personnel. But urbanization is a social cultural factor that can be make it easy for a first mover to reach the consumer. A first‐mover is also negatively influenced by legal political factors in emerging markets. Emerging markets can have less political stability, trademark protection and the government can implement investment restrictions. But there are also legal and political factors that are positive for first movers,  many emerging markets are forming trade agreements and there is liberalization in many industries. Lastly there are also competitive/marketing conditions there that are advantageous for a first mover. There is pent up demand for foreign products partly because local product are of lower quality. But first‐movers can also have a big disadvantage against local competition  because the consumers can prefer cheaper products or not be able to buy expensive ones.   . Page 14 of 62 .

(15) Figure 2: Impact of emerging market factors on first mover advantages (Nakata and Sivakumar 1992 page 479). The research by Nakata and Sivakumar (1997) was just conceptual, no actual cases or data were used. This thesis will add to the knowledge from Nakata and Sivakumar (1997) by analyzing an actual situation in which a first mover and a second mover compete in a developed and an emerging market. This can provide an empirical validity to the theoretical framework provided by Nakata and Sivakumar (1997) paper and give more detail about first and late mover advantages in different markets than what  a purely conceptual paper can give. It is also important to know what . Page 15 of 62 .

(16) elements of the Kerin et al (1992) model played a role in determining pioneer and follower advantages in different markets and how they were impacted by the market. Because this will give a more complete picture of what causes these advantages in various markets. We currently do not know if a company can easily expand its first mover advantage to a foreign country or if it will be easy for a follower to have an advantage there. According to the resource based view companies have different resources and capabilities. In the RBV companies compete with each other to obtain costly to copy resources. According to Barney (1991) these resources should hold VRIN values (valuable, rare, inimitable and not substitutable). Peteraf (1993) agreed that resources should be inimitable but added to this that resources should also have limits to ex ante and ex post competition and be immobile. According to Kerin et (1992) the size of first mover advantages is determined by economic factors, preemption factors, technological factors and behavioral factors. And late entrant advantages are caused by imitation costs, free rider effects, scope economies and learning from a first mover mistakes. Because markets differ the importance of these factors will also be different. The final outcome of the advantage is also impacted by the environmental factors in these markets (Kerin et al 1992). Kerin et al (1992) think that a company can only profit from a first mover advantage if it has the right resources and capabilities for the market opportunity. Prior presence in a market with different products can be part of these resources and capabilities. Kerin et al (1992) proposed that company strategy can impact the size of the first or late mover advantage. Strategy determines how the company wants to behave in relation to the market. The strategy that a company chooses is impacted by the expectations that a company has. This expectation can be related to the fit that a company has to the market taking into account the factors in emerging markets found by Nakata and Sivakumar (1992)  The fit between between a company's resources, capability and strategy can . Page 16 of 62 .

(17) differ between a developed and emerging market country because companies have different resources and capabilities in relation to markets with different characteristics. . RQ2: How are first and late mover advantages affected by the differences between a developed and emerging market? . Proposition 2: First mover and follower advantages are impacted by the strategy, resources and capabilities a company has in relation to the developed and the emerging market, based on the factors found by Kerin et al (1992), impacted by the Nakata and Sivakumar (1992) factors in an emerging market   . . Page 17 of 62 .

(18) 3. Methodology In this chapter I will describe the methods used to answer my research questions. I will also discuss why these methods were chosen. This section is important to give the reader confidence in my findings. Describing the methods used also makes it possible for other researchers to replicate my research findings. I will show what data sources and research methods were used. I will explain what cases were selected and on what basis they were selected. Also I will describe in detail the steps taken to analyze my results. . 3.1 Research design My thesis is both exploratory and descriptive. My thesis is exploratory because not much research has been done on resource and market differences relating to first mover (dis)advantage. Qualitative research fits well with exploratory research because its more broad and holistic than quantitative research. Qualitative research is more about understanding a problem than measuring a problem with numbers (Yin, 2003). Quantitative research is often used in later research stages after qualitative research to measure and prove causal relations. My research is also descriptive. I am describing how first mover products differ from follower products in the use of complementary resources used to commercialize products. I am also describing how first mover and late mover advantages are affected by the difference between developed and an emerging markets. Qualitative research methods also fit well with a descriptive research. With qualitative research its possible to describe phenomena in detail from multiple angles without having to rely only on measurable data (Yin, 2003).  . Page 18 of 62 .

(19) According to Yin (2003) the use of research methods should fit with the research questions. My research questions fit well with the use of multiple case study methods. Case study methods involves researching contemporary issues in their natural context (Yin, 2003). Case studies should be used to answer why and how questions (Yin, 2003) as opposed to how much or how many questions, there quantitative methods are more appropriate. My research questions fit these criterion. Case study methods are appropriate when the researcher has little influence over the events studied (Yin, 2003). This is the case with my research questions. It is good to use case study methods when little research has been done on a topic,  because the research questions can be answered using case study data (Eisenhardt, 1989), so there is less need to rely only on prior research.. Case study methods are appropriate to examine complex issues where the causes of issues and the relation with its environment are not clear (Yin, 2003).  The case study cases have to be seen in relation to their environment. This gives rich detail into the subjects studied.   . 3.2 Case selection All the case companies in this thesis are selected in the hardware and software computer technology sector. This will make the external validity low but the internal validity will be high. This means that the thesis will not give information applicable to all companies but it can give good information relating to the computer technology sector. It is important to focus on internal validity before external validity to create good qualitative research (Gibbert & Ruigrok, 2010). Also it can be expected because its a dynamic sector that first movers and later movers are important phenomena in the technology sector. The technology industry is very dynamic because many new technologies and products are introduced in it. The selected cases are all related to big well known public companies. This means it will be easy to find relevant information partly because public companies are obligated to disclose important information to the investor and many news sources . Page 19 of 62 .

(20) cover them. This is helpful because my data collection involves only secondary sources, these sources will be described further later in this chapter.    For my first research question regarding complementary resource differences I selected two successful well‐known first mover and two fast follower product lines, all the selected cases are described in the next chapter. A first mover product is new entry to the market (Lieberman and Montgomery, 1988). A fast follower is not the first to come to market but is able to move quickly after the first mover. This means it is one of the first to copy the innovator.   Using four cases has multiple advantages. First of all it can be a problem to have few cases because the findings can be to idiosyncratic or even random (Yin, 2003). This problem is particularly big with a single case study (Yin, 2013). Using four cases enabled me to use two kinds of replication logic to analyze the cases. Literal replication is used to see if similar selected cases have similar characteristics. Theoretical replication is used to see if differently selected cases have contrasting characteristics (Yin, 2003).   For the second research question about emerging and developed market differences I selected two of these product lines and added a developed and an emerging market embedded case to each of these cases. An embedded case study is a case study containing more than one sub‐unit of analysis (Yin, 2003). The identification of sub‐units allows for a more detailed level of inquiry.  For this research question analysis of the cases was made using only theoretical replication because there are no similar cases were used. This means I expect contrasting findings. To use this replication logic the cases were selected using theoretical sampling, the goal of   theoretical sampling is to choose cases which are likely to replicate or extend the emergent theory Eisenhardt (1989).  This type of sampling has also been called “polar types” by Eisenhardt (2007).   . Page 20 of 62 .

(21) 3.3 Selected Cases In this chapter I will present the selected cases. For my research question regarding complementary resources I selected the Apple iPhone and the Dell direct PC model as first mover product lines, because they were first to the market and very successful. I selected Samsung Galaxy and Microsoft Windows as fast follower product lines because they were very successful in following a first mover. For my research question regarding emerging versus developed markets I added the USA and China as embedded cases to the iPhone and Galaxy product line, because they are good examples of an developed and an emerging market.  I will in this chapter also show why my were selected. 3.3.1 RQ1 Selected cases Successful first mover: Apple iPhone. Apple introduced the iPhone in 2007 (FT, July 17 2007). The product was promoted as the combination of the iPod with a phone. The product ran on the easy to use iOS with popular apps like iTunes, iOS previously named iPhone OS is an operating system specifically designed for mobile touch screen devices. One of the biggest innovations of the iPhone was its touch screen, similar to the latest iPod released just before the iPhone, a keyboard was no longer needed. Although the iPhone is not the first smart‐phone it can still be considered a first mover product because it offered a totally new value proposition to the consumer. This product has currently still the number one position in the USA (comScore, March 7 2014). Because of its popularity the product was imitated by many other companies. . Page 21 of 62 .

(22) Successful first mover: Dell Turbo PC and the direct PC business model. Like Apple with the smart‐phone, Dell did not create the first PC. The computer Dell made was a first mover product because Dell had a new different business‐model than their competitors. Customers could directly order a computer from Dell, it cut out the middlemen. The customer could could also select customized product features (used components). The first computer Dell made in 1985 was called the Turbo PC. Dell did not create their own components like most computer companies did at the time but used virtual integration (Magretta, 1998) to create their products. This means the product was custom assembled with parts from third parties. With this business model Dell was able to grow from a small company to a global player in a few years. Successful fast follower: Samsung Galaxy. The first Samsung Galaxy (GT‐I7500) was released in 2009. At that time similar phones were also released under other names. I count other early Samsung models with Android OS and touch screen as part of its Galaxy product line. When it became clear that the iPhone was very successful Samsung was able to quickly follow it with a similar product. This has caused a patent battle that was largely won by Apple in the USA, but the fines Samsung must pay are small for such a big company. The Galaxy had features that are very similar to the iPhone. It had similar software with Apps called Android, made by Google. It also had a touch screen. And according to the Apple infringement case it had a similar look and feel. Samsung is seen as Apple's biggest competitors and it sells more smart‐phones in China (IDC February 2014), the biggest smart‐phone market in the world. . Page 22 of 62 .

(23) Successful fast follower: Microsoft Windows. Apple was a first mover when it created a revolutionary new operating system for the Lisa and Macintosh computer. In December 1979 Steve Jobs visited Xerox PARC where he first saw the technology for the mouse and graphical user interface in the never commercialized Xerox Alto computer (Isaacson, 2011). Xerox PARC can not be seen as the first mover because the alto was not commercialized it was only used internally by Xerox and at a few universities. In 1984 Microsoft made software for Apple on the Macintosh. In 1985 Microsoft introduced Windows an operating system with graphical user interface. This was very similar to the software created by Apple. The Windows follower product made Microsoft the biggest software company in the world. This case is a bit different as the other three cases because it is in the software and not hardware industry. But its still part of the computer industry, it will be interesting to see if my proposition hold for this case. 3.3.2 RQ2 Selected Cases Apple iPhone as first mover in the USA.. Apple introduced the iPhone in the USA in 2007 (FT, July 4 2007). The united states is a good example of a developed market. The mean income per person is high but growth is slow. The USA is an important smart‐phone market, it was the biggest market until it was overtaken by China in 2013 (flurry.com, February 18 2013). . Page 23 of 62 .

(24) Samsung as fast follower in the USA.. Samsung started selling its first Android based phones in 2009 including the first Samsung Galaxy (GT‐I7500). This phone is similar to the Galaxy product line because of the Android software and touch screen. Android was developed by Google in a consortium that included Samsung as an answer to the popular iOS. Samsung moved later than Apple but was fast as follower.   Apple iPhone as first mover in China.. Apple was also a first mover in China compared to Samsung. The first Apple store opened in Beijing in 2008 (businessweek.com, June 12 2012). Apple started selling the iPhone with China Unicom in 2009 (FT, September 29 2009 Asia edition). China is a good example of an emerging market with high growth but still a small mean income per person. China is the biggest smart‐phone market in the world since 2013 (flurry.com, February 18 2013). Samsung as fast follower in China.. Samsung started to sell its GT‐i6500U Saturn, also known as Galaxy. (FT, January 19 2010) with China Unicom in beginning 2010. This phone has Android 2.1 Eclair software and a touch screen. So it is very similar to what became the line of Galaxy smart‐phones.   . Page 24 of 62 .

(25) 3.4 Data sources Only secondary data sources were used. These include newspaper articles, company documents and third party interviews.  My research questions could be answered well with secondary data because there is very much information available on big public international companies. And my research questions did not demand detailed insider information.   As one source I used corporate websites and investor relations websites, they gave information regarding company history and product introductions. One problem with this source is that its hard to tell if the company is objective. This is why I used Financial Times newspaper articles accessed with LexisNexis academic and other news sources as most important source. Financial Times has a very good reputation and LexisNexis makes it possible to easily find articles that were written many years ago. Because its a third party this source can be very objective about companies. Only reputable news sources were used to ensure quality. Lastly some of the articles and documents used did include interviews done by third parties. Magretta (1998) for her article interviewed Michael Dell. Isaacson (2011) for his authorized biography interviewed many people related to Apple including Steve Jobs. News articles used also included interviews with Samsung executives. According to Yin (2003) it is best to use as many sources and methods as possible to make triangulation possible. Triangulation means the comparing of various data methods and sources. Although only secondary sources data were used triangulation was possible because I used different sources. . Page 25 of 62 .

(26) 3.5 Data Analysis In this chapter I will show what analytical steps were taken during the research. According to Yin (2003) its important to create a case study protocol.  This protocol is a mental framework guiding the research. A case study protocol can enhance quality and transparency. By showing the steps taken I also give a chain of evidence for my findings (Yin, 2003). Because qualitative research is iterative the steps shown were not followed in exact sequence as in a quantitative study. There was also no exact date when data was collected like in a quantitative research. The step are also summarized in a table. 1. My research started with a deductive approach, I researched current knowledge on first‐ mover (dis)advantage literature and based my research questions on gaps in this knowledge. After this also based on current research I created propositions about what I expected to find in the cases about my research questions. According to Yin (2003)  theory development prior to the collection of any case study data is an essential step. 2. Then I took a inductive approach and looked for successful product lines that were either a first mover or fast follower. After selecting four cases I selected embedded country cases for two of the product lines using theoretical sampling between developed and emerging markets. Theoretical sampling means purposely selecting cases to find similar or contrasting results (Eisenhardt 1989). 3. After the cases were selected I started to collect data on them using the data sources mentioned above. I looked for data on my cases using the research questions and propositions. For RQ1 (about complementary resources) this means the most important variable was who produced the components that together makeup the product. The data found per product line can be seen as a sort of product tear down, showing what parts . Page 26 of 62 .

(27) constitute the product. Researching product architecture is important because product architecture is a determining factor for industry architecture (Pisano and Teece 2007). For RQ2 (about emerging markets) the most important variables where the two models described in the literature section.  LexisNexis and Google where searched with keywords based on my propositions. All relevant down‐loadable articles were put in a database for later retrieval and analysis. Some articles were only available online, so links of these articles were stored in a browser. Yin (2003) advises creating a case study database. Data was collected until knowledge saturation. After the collection the data set consisted of 73 articles.   4. All articles were read to form a first impression on the outcome of my research questions. Articles that were not relevant or had outdated information were removed from the data set. 4 articles were removed so 69 remained. 5. Stored articles and web‐content then received codes for related chunks of data to enhance pattern matching, they can be seen as labels. These codes include the search keywords used and where created at first in a deductive way based on the research questions and propositions. This coding in categories is also called selective coding. An example of this kind of coding is the name of product lines e.g Samsung Galaxy that came from my case selection. The codes where revised inductively because of new information coming from my articles about the cases using open and axial coding, to find and categorize the new information. An example of this kind of coding is the distinction I made between internal and external complementary resources. My coding strategy was more deductive than pure grounded theory (Glaser and Strauss, 1967), which starts with open coding and than axial and selective coding. Because I have put emphasis on the theories that came from my literature section. The coding was done with Nvivo 10 the coding scheme is in the . Page 27 of 62 .

(28) appendix. This software was selected because of its reputation as a high quality and easy to use software product. Also Nvivo 10 has Ncapture, this mean that web‐content can directly be captured to be coded. For each of the two research questions a separate project file was made in the software to make the case studies easier to manage. A separate file was also made for the Tesla Motors case (see limitation section). This coding was useful to connect the data with the results what Yin (2003) calls creating a chain of evidence. I used within and cross case pattern matching to analyze my cases (Yin, 2003). This means that I for example looked at the complementary resources that were used to create the Apple iPhone and then compared this to the complementary resources that were needed to create the Samsung Galaxy phone. During the coding process I found some articles that had outdated news information or were less relevant to my research question as first tough. 10 articles articles were removed so 59 remained. 6. To help analyze the results the most important variables where put in tables that were later put in the result section. I have also looked for opportunities to find information based on the number of times a code came up in Nvivo. I found there were more references for external complementary resources for first mover product lines. While there were more references for internal complementary resources for follower product lines (see table 3.) 7. A narrative was written 8. Lastly there is a deductive stage where I compare my finding to current literature. . Page 28 of 62 .

No results found