Master Thesis
Submitted: 02/02/2017
Supervisor: Dr. James Slevin
In a smart connected product proposition, what factor will influence consumers’ attitude of software based capabilities, and how the software based capabilities function to consumers?
Words Count: 9018
Student ID: 10919155
Student Name: Jiabao Li
Acknowledgement
I would like to express my appreciation to my thesis supervisor Dr. James Slevin for his professional insights, patience and support on the completion of my thesis. During the thesis composition period, he was always supportive on giving his timely feedback. Guiding me on overcoming all possible barriers I occurred along the way. I am also thankful that he gave full support on my internship in order to complete the thesis.
With respect to the company Philips, I would also like to thank my manager whom I have been working with: Viola Wajer Huenges, for involving me in the project and providing insight of industry. She also gave me advice and support on data collection procedure.
I would also like to thank my friends who share their opinions and advice with me in terms of the thesis. Especially during the stressful time of combining thesis writing and internship, they always encourage me, keep me accompany and are willing to help. Special thanks to my friend Mike Nijenkamp, whom I can always count on in terms of suggestions, problem-solving and advice during the past few months.
Jiabao Li
3
Abstract
This study focuses on smart, connected products and elaborates on its three components: physical component, smart component and connectivity component. To answer the research question on how smart components being influenced by other factors and how it functions in a smart, connected proposition. The study draws attention on smart component and analyzes how three components interact with each other based on consumers’ attitude. To be specifically, the result presents that if consumers perceive the smart component itself is easy to use and therefore it is useful and meaningful, they would tend to like the smart component. Furthermore, if the previously existing physical component already establishes a good reputation in consumers’ mind, the newly invented smart component connected to it will also benefit. However, the research does not find a relation between connectivity and smart component, this may due to the limitation of case incorporated in the study. The research also finds with one special smart component: mobile coaching application would also function even applies it in a smart, connected product. The findings may be useful and show direction for future research on smart component and for companies who would like to invest on developing smart component and adapts it to existing physical product.
Keywords: smart, connected product, digital innovation, technology acceptance model, consumers' attitude.
Contents
Introduction ... 5
Theoretical Background ... 7
Smart, connected product ... 7
Information-Oriented Mobile Applications ... 8
Mobile coaching application ... 9
Technology Acceptance Model ... 10
Brand Extension ... 14 Case Description ... 15 Methodology ... 19 Participants ... 19 Research Design ... 20 Measures ... 22 Results ... 25 Data preparation ... 25 Testing Hypotheses ... 26 Discussion ... 32 Conclusion ... 35 Academic implication ... 35 Practical Implication ... 36
Limitation and Future Research ... 36
Appendix ... 39
Reference ... 39
Syntax of Main Analysis ... 44
5
Introduction
Nowadays technology has brought convenience to our life and keeps upgrading it to next level. With the help of smart home technology, you can monitor own house status via your smartphone; at the same time using the remote control to turn on air-conditioner, lights and TV. With embedded technology on your clothes, your Smartphone is able to tell you how much calories you have burned today, at the same time your connected car alert you about traffic jam and provide alternative fastest route; you may only need to hit a button on your phone again and your smart vacuum machine would start working at home. All the convenience we are benefiting is thanks to information technology. The first Information technology wave occurred during the 1960s and 1970s and dramatically influenced company strategies (Porter & Heppelmann, 2014). Information technology enables data recognition equipment, communications technologies and factory automation, so the values company makes, depends on how much customers are willing to pay for product or service (Porter and Millar, 1985). The second IT wave was caused by the prevalence of Internet technology between the 1980s and 1990s. The Internet has changed the whole industry structure: expanding the geographic market and enabling a better cooperation with suppliers and individual customers (Porter, 2001).
However, these two IT facilitated development did not change products selves, Toffler (1990) indicates in his book “The Third Wave”, by the start of 2000, there is a third wave enabled by information technology which brings us to Information Age. The booming of e-business allows intangible digital products such as e-books, music, video, software, and other intellectual property to step into the market. IT is becoming part of the product, digital sensor, software are embedded into the product; connect to a product cloud where all the data has been detected through the product will be stored (Porter & Heppelmann, 2014).
Moreover, these IT embedded products are called as Internet of things (IoT) products. The concept of IoT means the physical entities that incorporate planted technology are both connected to the Internet and able to communicate with other devices (Navales & Haslehurst, 2015). IoT phrase reflects the growing number of smart, connected products and highlights the new opportunities they can represent. Smart, connected products here refer to products embedded with processors, sensors, software and connectivity that allow data to be exchanged between the product and its surrounding, manufacturer, operator/user, and other products and systems. An example of smart, connected products is Volvo’s connected car. By implanting a Bluetooth technology in the car, users are allowed to open and start their car with a mobile application, share a key with friends, family and colleagues, and use car sharing services. Connectivity also enables some capabilities of the product to exist beyond the physical device. The data collected from these products can then be analyzed to advise decision-making, allow practical efficiencies and continuously enhance the performance of the product.
Currently, there are 5 billion connected products in the world not including computer, phones and tablets. By the year of 2019, this number is predicted to increase to 20-25 billion (Business Insider, 2015). As can be seen, many companies’ product innovation strategy will be embracing information technology and turning existed physical product into smart, connected products. By accommodating digital innovation into product innovation, the strategy will help companies in the new product development. Because a key feature of digital technology is the embodiment of software-based capacities into objects that previously had only physical materiality. Previous research regard to smart, connected products are either standing on a strategic or organizational level: how the smart products shape the new competition among companies and how to upgrade organizational innovation (Porter &
7 Heppelmann, 2014). However, there is not much literature analyzing software technology in the smart product proposition other than a technical view. Moreover, consumers' satisfaction on the smart product still needs to be improved. For example, a research done by Affinova (2015) shows that 41% of people are still reticent about the smart product: the products they have seen or heard of are simply gimmicky. How to build users' affection and satisfaction of the smart product? This research will adapt from consumers/marketing angle, test if consumers perceive software of connected products fit with previous communication from the company; explore if the software-based function has delivered expected communications to consumers and answer the question:
What factor will influence consumers’ attitude of software based capabilities in a smart connected product proposition and how the software based capabilities function to consumers?
By incorporating Philips’ new product development of smart connected shaver as a case study, the research will take a look at users’ feedback on home usage test and answer the research question. It is also more meaningful to choose big companies as an example because big enterprises normally have a mature routine of product development (Nelson & Winter, 1982). These routines mean first in operation level, everyone knows his or her task while knows limit about others; Second, the routines or practice are hard to change and are kept from changing by managers (Dougherty, 1992). It is understandable that for big firms new product innovation is slower and less effective. This paper may help on providing some insights and practical experience for large corporations who would like to develop connected products.
Theoretical Background Smart, connected product
The new risks are appearing with the development of connected product industry, especially for big companies. To illustrate, the barriers for new entrants are going down, while current competitive players are hesitant on fully adopt connected products, because they may prefer to protect present hardware and profitable legacy. New competitors such as companies lack off tangible products company may find new opportunities, incumbents are facing the situation of more and more substitutes coming up. Accordingly, it is essential for players in the market to have a better understanding of smart connected products. As previously mentioned, smart, connected product consists of three components: physical components, smart components and connectivity components (Porter & Heppelmann, 2014). Physical components represent the electrical and mechanical sectors of the product; Smart components are the parts that enable users interacting with the physical components, which can be sensors, software and data storages; Smart components will extend or amplify the value of the physical components; Connectivity component means the port or receiver which will enable the connections to the product; Connectivity component extends the value of smart component and even allows some of the values exist outside of the physical product. This research will focus on smart component and examine its relationship with other components (physical component and connectivity component). Because by taking advantage of the connectivity to a physical product, the smart component helps smart, connected product broaden the value proposition for both consumer and business side beyond product per se.
Information-Oriented Mobile Applications
Information-Oriented Mobile Applications (IOMA) can be considered as a practical example for the smart component. According to Chen, Meservy and Gillenson (2012), IOMA is defined as mobile applications that offer users timely, personalized, and/or localized information anytime; and anywhere on mobile devices. For example google maps, journey
9 planner app or cooking recipe app. It is predicted to increase notably in the coming years as mobile data services become prevalent and mobile providers open up their platforms to third-party developed applications (Malhotra & Segars, 2005). The usage of IOMA is depended on the use of a smartphone that is connected to the mobile Internet or wireless network. In most scenarios, an application is downloaded and installed on the smartphone, offering the user with the user interface and icon connect to the application content. Demanded information is transferred to the user’s device via the Internet. To conclude, for smart connected products, IOMA can take the role of the smart component, by analyzing the data generated from physical component and providing user relevant information based on their interest or need.
Mobile coaching application
Mobile coaching application is categorized into IOMA, it is a mobile application which offers users information and helps them achieve a certain goal or changes their behavior. Coaching is a type of improvement in which a learner or client is supported in accomplishing a particular personal or professional goal by provided training, advice and guidance. The goal of coaching is developing skills (Cacho-Elizondo & Shahidi, 2015). Different from mentoring, coaching focuses on specific tasks or goals, but mentoring talks about general objectives or overall development (Renton, 2009). By changing behavior to a healthier and more organized lifestyle, for return, people also experience an increase in self-efficacy. Successive mastery over a certain task helps people to develop a skill, or foster improvement on the coping mechanism for dealing with difficulties in life (Marcus, Selby, Niaura & Rossi, 1992).
The method of using certain technology to persuade people on changing behavior is persuasive technology. Persuasive technology is widely described as technology that is meant to have an influence on users' attitudes or behaviors via advice and social influence, but not
through force (Bailenson, Beall, Loomis, Blascovich, & Turk, 2004). Such technologies are normally adopted in sales, politics, military training, and management, and may possibly be implemented in any area of human-human or human-computer interaction. Most persuasive technology including desktop computers, Internet services, video games, and mobile devices. Fogg (2002) affirms that mobile devices are especially applicable for supporting the change of behavior, because they can intervene at the suitable time, in the fitting context and in a convenient way (as mobile device is always on); also assist social intercommunication, which fits in people’s interest in nowadays. Because people always carry them around, they become part of users’ personal life, people have a favorable attitude on using mobile device, so users are more likely to accept and follow the information it provides. To sum up, a mobile device is suitable for people who aim at changing behavior, which also falls in the interest of “coaching”; mobile coaching application can be acknowledged as a sufficient tool to help people realizing a specific goal. In the case of smart connected product proposition, mobile coaching application (smart component) works with physical component through connectivity component will possibly change users’ behavior. To generalize the above statements, we can conclude the first hypothesis:
H1: Users’ positive attitude on the smart component (mobile coaching application) positively influence device’s persuasive power on users (users are more likely to change their behavior).
Technology Acceptance Model
After identifying how smart component works on users actions. This research will define what factors affect users’ attitude towards smart component of a connected product, we can first delve into and see how consumers perceive the smart component’s performance and if it has an impact on people’s attitude. It has been proved that consumers’ attitude on new information has a crucial impact on successful information system usage (Davis, 1989). If the
11 information system adaptation is low; The company does not benefit from it, because in this way the company cannot learn from the data generated by users and improve the user experience. According to Pikkarainen, Karjaluoto and Pahnila (2004), users’ satisfaction of the information system (new technology) and usage of the new technology are considered as an emotional and cognitive measurement of users' adaptation.
One of the most common models used for analyzing technology adaptation is the technology acceptance model (TAM). The model was initially developed by Davis (1989) and being adjusted by other scholars, Davis (1989) introduces two dominant beliefs which are most influential to technology acceptance: perceived usefulness (PU) and perceived ease of use (PEOU); PU and PEOU relate to the attitude of use, which will affect use intention and finally to adaptation. Perceived usefulness, according to Davis (1989), means to what extent a user believes using a specific technology will improve his or her job performance, which has impact on attitude and intention to adopt. Perceived ease of use, is described as a user considers that understanding and using particular technology is effortless. Davis believes it influences PU ad also directly influences users’ attitude of the technology.
TAM is based on the theory of reasoned action (TRA) (Fishbein & Ajzen, 1975). Theory of Reasoned Action indicates that a person's behavior is decided by his/her intention to execute the behavior and that this intention follows by his/her attitude regards to the behavior and the person’s subjective norm. Although TRA and TAM shared similar root, there are also substantial differences that can be detected. First, TRA believes different factors that have effect on technology are highly related to different context hence they cannot be concluded. But TAM asserts two beliefs: PU and PEOU have an impact on all technology adoption (Pikkarainen, Karjaluoto and Pahnila, 2004). Moreover, TRA does not specify the influencing
factors on users ’attitude but TAM states PU and PEOU are two individual factors, with distinctive features, in this way it is easier to identify and analyze the affecting factors (Davis, 1989).
Combining with the “technology” this research focuses on, which is the mobile coaching application within the smart connected product, the following hypotheses are generated:
H2a: Users’ perceived usefulness (PU) of the smart component has a direct positive effect on users’ attitude on smart component (mobile coaching application).
H2b: Users’ perceived ease of use (PEOU) of the smart component has a direct positive effect on users’ attitude on smart component (mobile coaching application).
Furthermore, according to a study conducted by Hong and Tam (2006), PEOU also has a positive influence on PU because PEOU has both a direct and indirect impact on users’ attitude. Therefore, two more hypotheses are formed:
H2c: Users' perceived ease of use (PEOU) of the smart component positively influence users' perceived usefulness.
H2d: Users’ perceived usefulness (PU) of the smart component mediates the effect of users’ perceived ease of use has on users’ attitude of the smart component.
Connectivity component has crucial function in the whole smart, connected product proposition, it helps connect the smart component and physical component by transferring data collected to data cloud for analysis and later for user to interact with. Two main purposes are offered by connectivity (Porter & Heppelmann, 2014): First, it helps product to communicate and exchange data with operating surroundings, for example, users, other
13 products or data cloud. Secondly, connectivity allows some functionalities of the whole smart, connected product happen or exist outside of the physical product, it assists in broadening values beyond the product itself. A good example of how connectivity serves to the smart, connected product is Ralph Laurent's Polo Tech shirt, a wearable technology which is able to detect users' real time heart-rate, breathing data and a daily view of calories burned and steps taken. All these functions were served by a stretchy band which snaps into the shirt. Embedded Bluetooth technology delivers the data received to connected iPhone or iPad. As can be seen, all the new functionalities that cannot be served by shirt are realized via the connection to a smart device.
Moreover, connectivity plays an important role on realizing connected products’ three main capability: Monitoring, control and optimization (Porter & Heppelmann, 2014). Smart connected product supports on monitoring products’ condition, operation and external environment through its sensors and sending users timely information; Connectivity helps transferring data from the sensor to smart device and help it to interpret the data to user friendly language. Smart and connected product can be controlled by users via remote control or control via smart component; when a command is sent to the smart device, connectivity will activate according to change on physical component. Because of monitoring and control, the smart connected product can be improved, optimized and personalized during usage. Therefore, only by connections, users are able to explore more functions from the smart component, we propose the following hypothesis:
H3: Users hold more favorable attitude on the smart component with an assistance of connectivity component comparing to the users who do not receive help from connectivity component.
Brand Extension
To explore how the physical component of a smart connected product has an impact on consumers' attitude towards smart component, brand extension is employed. Brand extension is firm's strategic term which happens when a current brand name is used to enter a different product category, it also means introducing a new product to market as extensions (Aaker & Keller, 1990). In the case of smart connected products, a mechanical component can be existing product, by implanting technology the existing product can be connected to new developed smart component and these components together constitute a whole smart product proposition.
The effect of brand extension stems from categorization theory. Consumers tend to make sense from the existing products and service, consumers construct categorical representation in their mind to organize and interpret information they receive about the products. Categorical representation is the knowledge that stored in consumers’ cognitive classification and will be used during categorization (Loken, Barsalou & Joiner, 2008). Categorization is the process when consumer assign a new product to consumer category using categorical representation (knowledge). Therefore consumers can understand and construct opinion about the product. In general, people tend to interpret new product by using prior information of a similar product.
Brand extension is attractive to firms and is effective on reducing new products failure rates because brand extension brings consumers brand name recognition, which the new product can benefit from (Aaker & Keller, 1990). In this way, consumers' relationship of and insight of an established brand will be reflected in the new product. In terms of smart, connected
15 product developed from existing product, the bond will be more solid as not the only opinion from the brand but also from the current product will be projected to a new element of the whole proposition. The last hypothesis of this research is:
H4: In a smart connected product proposition, users’ attitude on previous existing physical component has a direct positive effect on smart component.
The conceptual model contains four hypotheses indicating how users' attitude on coaching app is
expected to be influenced by physical and connectivity component. Moreover, its own ease of use and perceived usefulness also have impact on users’ attitude. People’s attitude on coaching
application will also influence the eventual behavior change.
Case Description
The case analyzed in this research will be Philips’ smart connected shaver “FaceSmart”, which is in product development phase and will be launched in China in 2017. The case is relevant for our study because as a big company who develops various electric products, by
Perceived ease of use Users’ attitude on coaching app Perceived Usefulness Attitude on physical component Frequency of measurement H1 H2a H2c H2b H2d Connectivity H3 H4 Figure 1.0
entering the connected product industry, it is essential to invest and examine on how to invent software based capabilities which consumers are satisfied with. Royal Philips is an assorted technology firm that pursues to upgrade people's lives through meaningful innovation in the areas of healthcare, consumer lifestyle and lighting. The company is a market leader in cardiac care, acute care and home healthcare, energy efficient lighting solutions and new lighting applications, and in shavers, grooming and oral care products (Philips.nl, n.d.) China is one of Philips’ most profitable markets and is still contributing stably to sales. According to Philips CEO Frans van Houten’s report of Q3 in 2016 (Kharpal, 2016), Philips just had absolutely fabulous sales growth in China and the other emerging geographies such as Southeast Asia with double-digit order and double-digit growth. Among all the selling goods, Men's personal care products boomed, particularly in China, with shavers and male groomers being popular favorites (PHYS, 2016).
Philips’ brand slogan “Innovation and You” defines Philips difference from competition, people and innovation are two pillars of the company’s communication strategy, creating advanced technology in line with what people really want and need. These are meaningful innovations that help people live well, stay healthy and enjoy life. By introducing “FaceSmart” to Chinese male consumers, Philips aims to help them upgrade and change their total face care routine. “FaceSmart”, as a smart connected shaver, also composed of three components (as shown in figure 2.0): Existing physical component: Philips Series 7000 shaver (with shaver head and facial brush as attachments) and new physical component: click-on skin measurement device; Smart component: Skincare coaching app. Connectivity component: Embedded Bluetooth technology which allows shaver to connect with mobile app.
Consumers will be able to use skin measurement device to test their absolute value of facial oiliness and hydration. The result and skincare advice will be shown in coaching app, by
17 connecting shaver with coaching app, the shaver’s mode (speed and motion) will be adjusted according to user’s skin type. Moreover, coaching application is able to track shaver’s running time per use. The skincare coaching app for iPhone tailors user’s skincare routine based on his needs. After doing the skin assessment and taking a skin measurement, the app gives you guidance on how to optimize user’s routine every single day, the coaching app encourages users to measure their skin condition every day to understand and take better care of their skin.
As this research will focus on smart component of a connected product, we can take a look at the detailed skincare coaching app function (figure 2.1). By opening the application, the user is able to access his own skin profile (figure 2.1E), check the recommended skincare routine tailored to his skin type or skin issues (figure 2.1F), or take a new skin measurement (detect hydration and oil value) (figure 2.1G). The user also receives a few new cards every day
during the first app opening about skincare knowledge, advice and suggestion (figure 2.1A), by swiping all the cards the last measurement result will be presented in the end. The user will be prompted to take new measurement, the more measurement the user takes, the more accurate skincare advice will be given. Furthermore, received card collections (figure 2.1D), shaving and cleansing duration (figure 2.1C) and skin measurement history (figure 2.1B) are all tracked in the coaching app.
19
Methodology Participants
The participants of this study are the same as the target group consumers of “FaceSmart” so the result of analysis is more applicable and relevant. The participants who have contributed to this are people who are 26-40 years old Chinese male who are into skincare and use at least 3 skincare products, live in metro cities in China, personal income more than 10 thousand RMB per month, iPhone 6, 6 PLUS or 6S users. Because Chinese male who fit into these standards want to stand out with their appearance, they are the first to use new products and technology, spending a lot of time and money on their appearance and enjoy the shaving and skincare process as well as the results. In the end, 109 Chinese male were selected with mean age 31.58 (SD=4.12)
Sampling choice adapted in this convenience sampling done by social media recruitment with help of a research agency. A convenience sampling, which is frequently used in social research, is one that is simply available to the researcher by virtue of its accessibility (Bryman, 2012). In general, convenience sampling is researchers select participants from the range he or she can reach. Compared to other sampling choices, convenience sampling saves more time and money. Social media recruitment is recruiting participants by benefiting social platforms as talent databases or for advertising. Social recruiting uses social media profiles, blogs and other Internet sites to find information on candidates (Clements, 2012). There are certain advantages of applying social media recruitment: First, social media platforms enable a researcher to engage with their target audience and easily identify whether they fit for research criteria or not. Second, social media will spread the information about sampling request so it breaks the geographic boundaries and enlarged the pool of participants. Third, social media recruitment is considered as trustworthy so it results in a higher response rate.
In terms of current research, research agency posted request with criteria on their social media account and actively reach out to social media users who are suitable for research participants. However, there are also drawbacks for social media recruitment, such as if researcher’s account is less visible (does not have many followers), and target audience range is limited. Or the posted request may give away credential information. So in this case, a research agency takes the role of recruiter to collect participants; also in this case, the majority of target audience who are intrigued by new technology are also considered as heavy internet users which fit into the sampling choice.
Research Design
An online completion survey is selected to be the data-collection methodology. The self-completion survey, according to Bryman (2012), is self-administered, normally has few open questions and has an easy to follow design. There are several advantages of online self-completion questionnaire: First, cost and time wise, it is cheaper and quicker for researcher to collect data. As this study’s participants are from three metro cities which are geographically widely dispersed, an online questionnaire can be sent via email and distributed in big quantities at the same time. Second, the absence of administrator exhibits social desirability bias participants may have due to administrators’ characteristics. Third, it is very convenient for respondents, they can complete the questionnaire when they want and at the speed that they want to go. However, questionnaire is not suitable for in-depth research, because the questions are fixed and it is difficult to ask additional questions, moreover, the low response rate can also be considered as one risk survey leads to.
In order to evaluate the total “FaceSmart” proposition, a home usage test with 109 participants in China (Beijing, Shanghai and Guangzhou) started from 9th of May, 2016. Participants
21 recruited received the required product test samples by the agency and they were instructed to download the skincare coaching app. Participants were not given any perform specific tasks during the test period but they were advised to use the product sample every day for cleansing their face and shaving. The duration of the home use test was 4 weeks. At the end of the test period, participants received emails on their individual email address, to fill-in an online self-completion survey. The product tested was incentivized to participants after they have completed the task.
For purpose of having a better understanding of how each component works in the whole propositions, 109 participants were separated into three groups (figure 2.2): Group 1 (37 respondents): Click on skin measurement device with connected shaver (current FaceSmart product); Group 2 (36 respondents): Hand-held Skin analyzer, connected shaver; Group 3 (36 respondents): Click on skin measurement device with non-connected shaver and a mobile coaching application only provides new skincare tips daily. Product development chose to have group 2 with the purpose of testing different types of skin measurement device and benefiting future design. In this research, hypothesis 1, 2 and 3 were tested among all three groups. Hypothesis 4 which is related to physical component were only tested with group 1 and 2 which users tried out connected shavers. Accordingly, the surveys designed for each group also have slight differences.
Figure 2.2
Measures
Measurement Frequency (dependent variable) refers to the frequency of skin measurement, the skin coaching app aims to raise users’ awareness of the skin measurement importance, because in this way they can keep an eye on how their facial skin changes every day; Their frequency of skin measurement is predicted to be higher when they have positive attitude on coaching app.
The frequency of skin measurement was examined with one question “How many times per week did you measure your skin?” Participants could indicate on a 5-point scale ranging from
did not measure at all, less than once, 1-2 times, 3-5 times to more than 5 times. The answers
were summed up and calculate the mean to form an index. The frequency people measure their skin (M= 3.34, SD= 0.66).
Users’ attitude on coaching app (independent and dependent variable) is measured by three questions within Q20, because Q20 is the only scale that measures participants’ attitude on coaching app as a whole, the three questions are asked on a 5-point Likert scale from strongly
disagree, disagree, neutral, agree to strongly agree. Three indications are: “The app offers
something new”, “This app will have a positive impact on my life” and “My expectations from the app are met”. The answers were summed up and calculated mean as a new variable Attapp_mean (M=3.91, SD=0.53). All three factors load on one component with Eigenvalue
23 above 1. The three-item scale has a Cronbach’s alpha value 0.75, therefore it is a reliable scale.
Perceived ease of use (mediator) is measured by 11-items Likert scale, the 11 items are extracted from 13-items within one question Q16 (M= 3.36, SD= .45), and these 13 items are adjusted from system usability scale (SUS). SUS consists of ten-item attitude Likert scale (figure 3.0) and giving a global view of subjective assessments of usability. It was developed by John Brooke in 1996, and it is proved to be a reliable, low-cost usability scale that can be applied for global assessments of systems usability (Brooke, 1996). There are three factors covered in the scale: effectiveness (the capacity of users to finish tasks using the system, and the quality of the result of those tasks); efficiency (level of resource expended in performing tasks) and satisfaction (users’ attitude of system). The adjusted version of SUS is also shown below (figure 3.1), there are 4 more items which are repetitive items for the same purpose, this research only selected the items relate to PEOU. The variable contains reliable scale (Cronbach’s alpha= .74).
Perceived usefulness (independent variable) is measured by one question Q18 "Can you tell us how useful was the advice given by the app to take care of your skin". This question contains two 5 points scale items from strongly disagree to strongly agree: "The app provided me with useful skin advice based on the skin measurement" and "The app provided me with the right skin cleansing routine". The two items were summed up and calculated mean (M=3.90，SD= .55). Correlation. The two items have a Cronbach Alpha= .700 which means this is a reliable scale.
Connectivity (independent variable) is measured between group 1 and group 2, as two groups have experienced same physical devices but one is with connected shaver and one is normal shaver. Users’ attitude on mobile application (M=3.91, SD= .53) will be compared between these two groups.
The attitude on physical component (independent variable) is measured by one question “Q6_2 How much do you like the electric shaver which is part of the FaceSmart proposition” (M=3.77, SD=0.62) on a 5 point Likert scale from strongly dislike, dislike, neutral, like to
strongly like.
Figure 3.0
25
Results Data preparation
In order to facilitate analysis, the current data was cleaned and recoded, as the total 109 participants all accomplished the survey and they were selected by criteria, so no case was deleted. Firstly a few frequency tests were run to understand the participants better. Frequency tests show that most of the participants (30.9%) monthly household income is more than 18000 RMB (2463 Euros) (M=8.49, SD=2.90). Most of the participants (62%) with university degree (M=3.66, SD=0.51). 74.5% of participants only use electric shavers, the others use both electric shavers and blade, 61.80% currently using Philips shavers. Which means Philips brand is most appreciated by the participants. 65.50% of 109 participants shave 3-4 times per week. All of them agree or strongly agree that they are always on the looking for the latest developments for personal care products and are happy to try new products. All of the respondents agree or strongly agree on hoping their shaving and cleansing routine to get more innovative and comfortable because their appearance is important to them. They are willing to spend more money and time on it. All these results further indicate that the recruited participants have a favorable attitude on skincare, or are willing to get engaged with more skincare knowledge.
In assist of analysis, Q15_2 skin measurement frequency is recoded to Q15 from scale 1-5 to scale 0-4, so the scale becomes continuous (M= 3.80, SD= .57). Three items in attitude on coaching app are recoded to Q20_1, Q20_2 and Q20_3 following the same rule, the three items result were summed up and calculated mean and formulated into one new variable Mean_Att_app (M= 2.90, SD= .53). Following the same rule, two items of perceived usefulness were recoded to Q18_1 and Q18_2, with one new variable Mean_PU (M= 2.90,
SD= .55). Users’ attitude on physical component is recoded to Q6_2re with scale from 0 to 4
(M= 3.77, SD= .62).
In terms of PEOU Q16, after recoded the scale from 1-5 to 0-4, a factor analysis is run to check if all of the items are still measuring the same variable as the current scale is adjusted from the original system usability scale. Firstly, Item Q16_ 2and Q16_7 are taken out from the scale as they are not directly measuring PEOU. Second Q16_3, Q16_5, Q16_7, Q16_9, Q16_11 and Q16_12 are measuring perceived difficulty of using coaching app; thus these scales are reversed for factor analysis.
A principal component analysis (PCA) is conducted with 11 items. The result shows these items are loading on two components with both Eigenvalue-criterion bigger than 1 (component 1=4.66, component 2=1.66); and the scree plot shows there are indeed two components. Together the two components explain 57.44% of the variance in the original variables (items). Moreover, the table below clearly shows reversed scales are loading on the same component. As the result shows two different factors which are not expected. This research splits two components to two variables with mean calculated Mean1_PEOU (consists of Q16_1, Q16_4, Q16_8, Q16_10, Q16_13) (M= 2.63, SD= .63) and Mean2_PEOU (consists of Q16_3, Q16_5, Q16_9, Q16_11 and Q16_12) (M= 2.00, SD= .76), both are reliable scales with Cronbach’s Alpha= .77 and .87).
Testing Hypotheses
According to the research model, for testing hypothesis 1, a simple linear regression was first calculated to predict the dependent variable skin measurement. Hypothesis 1 states that the more positive attitude users hold for the mobile skincare coaching application, the more often
27 they will use the skin measurement device. Per this hypothesis, independent variable is users’ attitude on skin coaching application, dependent variable is skin measurement frequency. The regression model is significant, F (1, 107) = 74.59, b= .96, b*= .64, p <.000 with an R2 of .41. The regression model can, therefore, be used to predict measurement frequency. With each unit increases of users’ attitude on skin coaching app, there is .69 unit increasing on users’ skin measurement frequency. Hence, Hypothesis 1 is supported.
Table 1
Regression result for users attitude on coaching application predicting skin measurement frequency
Variable B SE B Beta T P
Attitude on coaching application 0.69 0.08 0.64 8.64 .00 Note R² = .47 (N = 109, p < .001)
According to the mediation model, a simple linear regression was first conducted to predict the dependent variable Users’ attitude on coaching app, based on the independent variable users’ perceived usefulness (hypothesis 2a). The result shows that the direct effect of perceived usefulness on users’ attitude on coaching app is statistically significant and positive
F (1, 107) = 30.03, b= .45, b*= .47, p <.000 with an R2 of .22, which means the model predicts 22% of variance. For every unit of increase in perceived usefulness, people’s attitude towards coaching app will increase by .69. Which means the more users believes coaching app is useful, the more positive attitude they will hold towards coaching app. Thus, hypothesis 2a is confirmed.
Table 2
Regression result for perceived usefulness on coaching application predicting users attitude on coaching application
Variable B SE B Beta T P
Perceived Usefulness 0.45 0.82 0.47 5.48 .000
Scenario 1:
As perceived ease of use now is divided into two variables: Mean1_PEOU and Mean2_PEOU, the mediation model is also analyzed in two scenarios accordingly. Scenario 1 adapts Mean1_PEOU as the variable perceived ease of use.
A linear regression model is followed to examine hypothesis 2b with perceived ease of use as independent variable, users’ attitude on coaching app as dependent variable. The result also presents a statistically significant relationship between perceived ease of use and users’ attitude on coaching app F (1, 107) = 4.40, b= .17, b*= .20, p <.05 with an R2 of .04, which means the model explains 4% of the variance. For every unit increase in perceived ease of use, the score of people's attitude on coaching app increases .45. The more users think coaching app is easy to use, the more positive attitude they will hold for coaching app. Therefore, hypothesis 2b is supported.
Table 3
Regression result for perceived ease of use on coaching application predicting users attitude on coaching application
Variable B SE B Beta T P
Perceived Ease of Use (Mean1_PEOU)
0.17 0.08 0.20 2.10 .038
Note R² = .04 (N = 109, p < .05)
In order to test hypothesis 2c, a simple linear regression model is conducted with perceived usefulness as dependent variable perceived ease of use as independent variable. There is a statistically significant relationship found between the two variables F (1, 107) = 6.31, b= .21,
b*= .24, p <.05 with an R2 of .056, which means the model explains 5.6% of variance of the model. With every unit increase on perceived ease of use, there is a .21 increase of perceived usefulness. When users feel the app is easy to use, they tend to also believe it is useful.
29 Table 4
Regression result for perceived ease of use on coaching application predicting perceived usefulness
Variable B SE B Beta T P
Perceived Ease of Use (Mean1_PEOU)
0.21 0.08 0.24 2.51 .01
Note R² = .056 (N = 109, p < .05)
To fulfill the mediation analysis, the above hypotheses (H2a, H2b and H2c) were conducted and all the hypotheses were supported. These results support the mediational hypothesis, thus the mediation analysis is able to continue. In the process of testing mediation, the research carried out a multiple regression while controlling the influence of perceived usefulness, with independent variable perceived ease of use and dependent variable users’ attitude on skincare coaching app. According to the result, perceived ease of use is no longer a significant predictor of users’ attitude on skincare coaching app after the mediator (perceived usefulness), consistent with full mediation (b= .17, p< .05; b’= .43, p= .000; Sobel’s Z= 2.50. p< .05). Approximately 23% of the variance in users’ attitude on skincare coaching app is accounted for the predictors (R2 = .23). The relationship between perceived ease of use and users’ attitude on skincare coaching app is fully caused (mediated) by users’ perceived usefulness. The effect of perceived ease of use on users’ attitude on skincare coaching app is indirect (runs through perceived usefulness) and will, therefore, disappear when controlling for perceived usefulness. Hypothesis H2d is supported. A full mediation effect is found.
Table 5
Regression result of the mediated relationship of perceived ease of use on users’ attitude on skincare coaching app through perceived usefulness
Variable B SE B Beta T P
Perceived Ease of Use (Mean1_PEOU)
0.08 0.08 0.09 1.07 .29
Perceived Usefulness 0.43 0.09 0.45 5.08 .000
Scenario 2
Scenario 1 adapts Mean2_PEOU as the variable perceived ease of use, the same procedure is followed as previous analysis.
A linear regression model is followed to examine hypothesis 2b with perceived ease of use as independent variable, users’ attitude on coaching app as dependent variable. The result also presents there is no statistically significant relationship between perceived ease of use and users’ attitude on coaching app F (1, 107) = 2.59, b= .11, b*= .15, p = .11, R2
= .02. Hypothesis 2b is rejected, there is no relationship between perceived ease of use and users’ attitude on skincare coaching app. As H2b is one important section to construct mediation analysis, in this way the mediation cannot be carried out. Hypothesis 2d is also rejected. Table 6
Regression result for perceived ease of use on coaching application predicting users attitude on coaching application
Variable B SE B Beta T P
Perceived Ease of Use (Mean2_PEOU)
0.11 0.14 0.15 1.61 .11
Note R² = .02 (N = 109, p > .05)
To sum up, there is a full mediation found using Mean1_PEOU as variable. There is no mediation found using Mean2_PEOU as variable
According to hypothesis 3, users who experience blue tooth connection between physical component and smart component have more positive attitude on skincare coaching app than users who do not experience the connection. In this case, we predict that group 1 and group 2 would have favorable attitude than group 3. There will not be a difference expected from group 1 and group 2. A one-way ANOVA was conducted among three groups to compare if
31 connectivity has an effect on users’ attitude. The result shows there is no significant difference as a whole F (2,106) = .06, p= .94, η2= .001. Therefore, hypothesis 3 is rejected.
Table 7
ANOVA Result with dependent variable users’ attitude on coaching app, independent variable as three groups with or without blue tooth connection
Variable Sum of Square df F P
Between Groups 0.03 2 0.06 .94
Within Groups 30.88 106
Total 30.91 108
Hypothesis 4 indicates that users who have a positive attitude on physical component (electric shaver) would also tend to have a favorable attitude on skincare coaching app. As only group 1 and group 2 users are using connected shaver, so only these two groups of people are taken as selected cases, group 3 in this case is excluded. A simple linear regression was calculated to predict users’ attitude on skincare coaching app based on users’ attitude on users’ attitude on connected electric shaver. A significant regression equation was found F (1,107) =192.41,
p< .000, with an R2 of .64. Participants’ predicted attitude on app is equal to .98 + .69 participants’ attitude on physical component. Participants’ attitude on app increased .69 for each unit increase on attitude with physical component. Therefore, there is a positive relationship between people’s attitude on connected shaver and people’s attitude on coaching app. Hypothesis 4 is supported.
Table 8
Regression result for users attitude on connecting shaver predicting attitude on coaching application
Variable B SE B Beta T P
Attitude on connected shaver 0.63 0.09 0.63 6.75 .00 Note R² = .39 (N = 73, p < .001)
Discussion
This research paper examined within a smart connected product proposition, to address the problem now in industry, which is hard for big firms to fully embrace new product innovations in terms of smart connected products. Moreover, to fill the gap of academic research which mainly focus on smart connected products in organizational and industrial level. The research communicates around software component within smart connected products and attempts to answer the specific research question: What factor will influence consumers’ attitude of software based capabilities? And how it will have an impact on consumers’ behavior. In general, we see very positive results which mainly aligns with research expectation.
The result of home visit survey provided support for the importance of physical component’s impact and confirms that smart component can also influence people’s behavior. As predicted by hypothesis 1, users’ attitude on the smart component (mobile coaching application) positively influence devices persuasive power on users (users are more likely to change their behavior). The research finding aligns with previous research about coaching app on the smartphone can assess freshmen undergraduates' time spent, in order to change their lifestyle and eventually improve their self-awareness (Runyan, Steenbergh and Bainbridge et al, 2013). The result also shows that high frequency app use and positive attitude towards app prompted greater self-awareness regarding time-management. Scholars suggest that app use may have caused some users to make changes in how they spent their time, therefore accelerating positive changes in their behavioral patterns. As can be seen, the behavioral change is not the ultimate goal of coaching app, to take a better control of own life can be achieved by changing behavior. In this case, by taking better care of facial appearance help users gain self-confidence and improve mental health.
33 In addition, hypothesis 2a is supported, if users perceive the information system as useful, they would generate a favorable attitude towards the information system. This finding is in line with the research of Zviran (2004), the researcher finds a strong correlation between perceived usefulness and user satisfaction under a different context within information system. The research also indicates that there are more sub-groups may have effect between this relationship for example participants' professional background, age and education level all subjectively influence their satisfaction level. Therefore our result is valid only within this certain participants.
Moreover, in scenario 1, hypothesis 2b and 2c are all confirmed, a mediation relation is found. Which means when users perceive high level of ease of use, they tend to feel the information system is useful, also they tend to hold a positive attitude towards the information system. The finding aligns with previous studies as well. Chau (1996) claims that ease of use has the largest influence on software acceptance; Igbaria, Zinatelli, Cragg and Cavaye (1997) indicate that perceived ease of use is a dominant factor in explaining perceived usefulness and system use; Hu et al. (1999) find out in their study that TAM model is able to provide a reasonable depiction of users’ adaptation on information system. Perceived usefulness was also found to be a string factor on determining users’ attitude on information system. After confirming hypothesis 2b and 2c, this research also discovers a full mediation, perceived usefulness mediates the effect from perceived ease of use on attitude towards software. Which means if people find the software is easy to use but if they perceive it is not useful anyway, they won’t be satisfied with software eventually. The logics have also been tested and confirmed by other researchers. Davis, Bagozzi and Warshaw (1989) conclude that usefulness and enjoyment together mediate the effect from ease of use on use intention. To sum up, TAM has been
examined to be a reliable theoretical model in understanding users’ behavior towards information system. It has been proven in certain empirical study and the model has been tested to yield statistically valid results. However, scenario 2 has rejected the relationship between perceived ease of use and attitude. The cause of two factors is examined, the participants in this research tend to agree, for example, “Q16_5 I think this app is designed for expert users” has a mean of 3.32 while “Q16_8 I am sure this app is easy to learn for new users” has a mean of 3.65, participants don’t tend to give a low score even if they don’t agree. This has to do with Chinese culture and people’s habit on answering survey.
In terms of hypothesis 3, it is expected that respondents who have experienced connected products are more satisfied with the mobile coaching application comparing to respondents who haven’t used connected product. This hypothesis was rejected by analysis. Although previous paper believes that when software is adequately integrated with micro-electronic sensor or mechanical technologies, the products will become accordingly smart, interconnected and adapted. Therefore the software can access and sense the external condition and is able to interact with other products and people, eventually changing according to what people need (Hebner, 2009). The present research manifests no significant evidence that people would be more content while benefiting from connectivity.
There might be several possible reasons, first root cause might be privacy issue. According to, consumers also have concerns of the probable negative effect of inviting more connected, smart products into their lives. Their personal data or privacy may be potentially transmitted to other devices or people beyond their decision (Cronin, 2010). The second reason might be because of the case study itself, as the connected shaver is still in initial developing phase. The enabled functions are still quite limited, current version allows users to personalize their
35 shavers based on their skin type. The shaving history will also be synced to coaching app. But besides the above, there are no instructions on how to shave to prevent shaving related issues for example irritation, redness and cuts. To sum up, there is no strong connection yet between physical component and smart component in this proposition. Thus consumers haven’t experienced expected multiple functions facilitated by Bluetooth connection, this might be the major reason that the analysis finds no significant difference among three groups.
With regard to hypothesis 4, the outcome supports the notion that the more people like physical component, the more affection they will hold for the smart component. Especially the physical component in previous existing products, which already established its reputation in consumers’ mind. The result paralleled with the study of Aaker and Keller (1990), according to their research, higher quality evaluation toward the original brand predict more positive attitudes towards the extension. Moreover, the research also points out that the relationship of a positive quality image for the existing product with the perception of brand extension is strong only under the condition that there is a basis “fit” between the two products. Hence in the current study, we can also infer that the mobile coaching application is also considered as “fit” with the image of Philips shaver, there is no big gap between two products in consumers’ opinion.
Conclusion Academic implication
Current Research surfaces three academic implications. First, instead of taking smart, connected product as a whole to analyze, present paper reviews separate component and examine how they influence each other. In this way, this research contributes a method to trace root cause of consumers’ attitude on connected product. Second, current research
provides a special case, which is also common to markets, the smart connected product is developed from existing product which will become physical component of the whole proposition. Smart component is the innovative component for connected product, by exploring how smart component interacts with other components offers new angles for academic study. Third, looking into one specific software: mobile coaching application, the outcome provides a certain context for analysis of coaching application.
Practical Implication
This study also provides empirical implications for corporation and connected product development. First, companies should invest on products which already have reputation among consumers, users’ trust on existing product will build a positive pre-perception of smart component. Second, companies should focus on consumer research and understand their digital usage habit, in order to design software in a way target users identify as easy to understand and valuable and eventually improves users’ satisfaction with smart component. Third, the developing team should set up key performance indication for smart component itself within connected product, based on feature of smart component. For example, coaching app aims at changing people's behavior and promote healthy lifestyle. Thus effectiveness of smart component can be tracked. The last, as connection, differentiate connected products with other similar competitors. It is crucial to introject on functions empowered by connection, if the functions are not taken as special and useful by users, connected products cannot highlight its advantages from all the other generic non-connected products.
Limitation and Future Research
Like most studies, the present study suffers from certain limitations which can be improved in future research. First, the sample size of this research is too small (109 participants), within 36
37 or 37 people in one group, although these testers all have been through four weeks usage, evaluations may still have the risk of low external validity. In the future research, recruiting more users would help to confirm and further explore the result. Second, during data analysis, we found that participants tend to agree with statements of questionnaire, which is a typical social desirability bias from Chinese respondents. Therefore our research’s internal validity is weakened and in previous result section, the study has to conduct two scenarios’ analysis. To avoid similar problems in the future, further studies should consider applying without potential statement, for example, instead of asking for rating “I think this app is easy to use”, the survey can be designed as “Please rate how easy you think the app is to use”. Third, result of the study is also limited by case, hypothesis 3 is rejected in this research which means consumers don’t perceive any difference of mobile coaching app no matter if it is connected to face shaver or not. However, as previously mentioned, as there are only a few functions enabled by Bluetooth connection in present case, there is a chance that users haven’t experienced the difference because of connection function is not fully developed. Future study direction is to investigate once more functions unlocked by connection, conducting the same study again and see if different result came up.
Moreover, due to the restriction of data, there are a few other interesting topics that can be further investigated. For example, this study talks about what factors influence smart component and within the smart component category. The research focuses on coaching application and examining if coaching application still has its impact on consumers once it becomes a component in a smart product proposition. While following up research can dig into how people’s perception of smart component influences their interpretation on the whole smart connected products. In addition, it would be interesting to allow the cultural heterogeneous as a factor in current model by testing consumers in different countries.
Furthermore, other types of smart components can be explored other than coaching application, it can be expected that as coaching application plays a crucial role in current case, other smart components which may only exist to assist physical component for a better performance. Nevertheless, it is important that more investigations can be taken place to analyze smart connected products from consumer point of view, which would help companies make decision on connected products development direction and eventually make a change on the industry.
39
Appendix Reference
Aaker, D. A., & Keller, K. L. (1990). Consumer evaluations of brand extensions.The Journal of Marketing, 27-41.
Affinova (2015), Innovation Trend Watch. The Internet of Things. Can it find a foothold with mainstream audiences today? http://iofthings.org/s/Internet_of_Things_Report__Web_.pdf
Bailenson, J. N., Beall, A. C., Loomis, J., Blascovich, J., & Turk, M. (2004). Transformed Social Interaction: Decoupling Representation from Behavior and Form in Collaborative Virtual Environments. Presence: Teleoperators & Virtual Environments, 13(4), 428-441.
Brooke, J. (1996). "SUS: a "quick and dirty" usability scale". In P. W. Jordan, B. Thomas, B. A. Weerdmeester, & A. L. McClelland. London: Taylor and Francis.
Becker, S., Bryman, A., & Ferguson, H. (2012). Understanding research for social policy and social work: themes, methods and approaches. Policy Press.
Cacho-Elizondo, S., Shahidi, N., & Tossan, V. Educational Mobile Coaching: Exploring Adoption Patterns and Barriers.
Chau, P. Y. (1996). An empirical investigation on factors affecting the acceptance of CASE by systems developers.Information & Management,30(6), 269-280.
Chen, L., Meservy, T. O., & Gillenson, M. (2012). Understanding information systems continuance for information-oriented mobile applications.Communications of the Association for Information Systems,30(9), 127-146.
Clements, A. (2012, Feb 16). Social media and recruitment: time for a coherent HR strategy? HR Magazine. Retrieved from http://www.hrmagazine.co.uk/article-details/social-media-and-recruitment-time-for-a-coherent-hr-strategy
Cronin, M. J. (2010).Smart products, smarter services: strategies for embedded control.
Cambridge University Press.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS quarterly, 319-340.
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer technology: a comparison of two theoretical models.Management science,35(8),
982-1003.
Dougherty, D. (1992). Interpretive barriers to successful product innovation in large firms. Organization science, 3(2), 179-202.
FishbeinM., and Ajzen I,(1975).Belief, attitude, intention and behaviour. Massachusetts:
Addison Wesley Publishing.
Fleck, L. (1979), Genesis and Development of a Scientific Fact, T. Trenn and R. K. Merton (Eds.), translated by F. Bradley and T. Trenn; originally published 1935; Chicago:
University of Chicago Press.
Fogg, B. J. (2002). Persuasive technology: using computers to change what we think and do.Ubiquity,2002(December), 5.
41 Gartner, Inc. (December 12, 2013) Gartner Says the Internet of Things Installed Base Will
Grow to 26 Billion Units By 2020. Retrieved from:
http://www.gartner.com/newsroom/id/2636073
Hebner, S. (2009). Smarter products: the building blocks for a smarter planet. IBM.
Hong S.J. and Tam K.Y. (2006). Understanding the Adoption of Multipurpose Information Appliances: The case of Mobile Data Services, Information Systems Research, 17, 2, June, 162-179.
Hu, P. J., Chau, P. Y., Sheng, O. R. L., & Tam, K. Y. (1999). Examining the technology acceptance model using physician acceptance of telemedicine technology.Journal of management information systems,16(2), 91-112.
Igbaria, M., Zinatelli, N., Cragg, P., & Cavaye, A. L. (1997). Personal computing acceptance factors in small firms: a structural equation model.MIS quarterly, 279-305.
Kharpal, A. (2016, Oct 24). Dutch health tech firm Philips sees Q3 profit beat with 'fabulous' growth in China. CNBC. Retrieved from http://www.cnbc.com/2016/10/24/dutch-lighting-firm-philips-reports-18-percent-increase-in-net-income-for-q3.html
Loken, B., Barsalou, L. W., & Joiner, C. (2008). Categorization theory and research in consumer psychology.Handbook of consumer psychology, 133-65.
Malhotra, A., & Segars, A. H. (2005). Investigating wireless web adoption patterns in the US.
Marcus, B. H., Selby, V. C., Niaura, R. S., & Rossi, J. S. (1992). Self-efficacy and the stages of exercise behavior change. Research quarterly for exercise and sport, 63(1), 60-66.
Nagy, I. (2013, Aug 29) The benefits of social media recruiting. Onrec The online recruitment resource. Retrieved from http://www.onrec.com/news/news-archive/the-benefits-of- social-media-recruiting
Navales, E., Haslehurst, R. The IoT Opportunity: Is Your Go-to-Market Strategy as Smart as Your Product? L.E.K. Consulting / Executive Insights Volume XVII, Issue 22
Philips Official website (n.d.) Retrieved November 12th, 2016 from Philips.nl: http://www.philips.nl/a-w/about-philips/bedrijfsprofiel.html
PHYS (2016, Jan 26). Electronics giant Philips posts 2015 profit up 55% (Update). PHYS.
Porter, M. E. (2001). OT COPY. Harvard business review, 1.
Porter, M. E., & Heppelmann, J. E. (2014). How smart, connected products are transforming competition. Harvard Business Review,92(11), 64-88.
Porter, M. E., & Millar, V. E. (1985). How information gives you competitive advantage.
Renton, Jane (2009). Coaching and Mentoring: What They Are and How to Make the Most of
Them. New York Retrieved from http://phys.org/news/2016-01-electronics-giant-philips-profit.html
