The focus of master’s thesis is to examine current possibilities of electronic marketing and means of their utilisation. The thesis analyses various methods of internet marketing and process of planning e-marketing strategy, including the possibilities of measuring its success.

This thesis deals with the creation of e-commerce companies annd prooposing solution to implement e-commrce. The theoretical part specifies the characteristics of electronic commerce, conditions and legal aspects of trading on the Internet. The practical part includes draft versions of e-commerce and its economic evaluation.

According to eMarketer, the total advertising spend in US alone was estimated to be over $238 billion. Firms invest large amounts of money in advertising to promote and inform consumers about their products and services, as well as to persuade them to purchase. The broad theme of advertising has been examined from many different angles in the marketing literature, ranging from empirically measuring effects of TV ads on sales to analytically characterizing the key economic forces stemming from enhanced targetability in online advertising. The purpose of my dissertation is to study some of the key questions which remain unaddressed in the advertising literature. In the first essay, I examine firms' choices of advertising content in a competitive setting. I demonstrate that competitive forces sometimes induces firms to choose advertising content that shifts consumers' perception of product quality. While this strategy hurts firms in a monopoly setting, it increases their profits under competition because it may increase the utility of their offering in comparison with the competing offering. In the second essay, I investigate the optimal mechanism for selling online ads in a learning environment. Specifically, I show that when ad sellers, such as Google, design their ad auctions, it is optimal for them to favor new advertisers in the auction in order to expedite learning their ad performance. In the third essay, I study the impact of tracking consumers' Internet activities on the online advertising ecosystem in the presence of regulations that, motivated by privacy concerns, endow consumers with the choice to have their online activity be tracked or not. I find that when ad effectiveness is intermediate, fewer ads are shown to opt-in consumers, who can be tracked and have their funnel stages inferred by advertisers, than to opt-out consumers, who cannot be tracked. In this case, consumers trade-off the benefit of seeing fewer ads by opting-in to tracking (positive instrumental value of privacy) with the disutility they feel from giving up their privacy (intrinsic cost of privacy). Overall, these findings shed light on novel strategic forces that provide guidance for marketers' advertising decisions in three distinct contexts.

Modern, technologically-enabled markets are disrupting many industry sectors, including transportation, labor, lodging, dating services and others. While the system operator is able to collect data and deploy various control levers, these systems are highly complex, marked by a large number of interacting self-interested agents, uncertainty about the future and imperfect demand predictions. There remain major challenges in optimizing these marketplaces. In this dissertation, I describe work designing novel algorithms and performing theoretical analysis of networked systems, including those that arise in marketplaces. I demonstrate how to use tools from applied probability, modern optimization, and economics to develop methodologies for online decision making in contexts such as queueing control, revenue management, and running a matching platform. The first part of the dissertation designs novel algorithms for dynamic assignment and revenue management. The work considers networked systems where agents or tasks arrive over time, which is broadly relevant to service platforms with heterogeneous services, for instance shared transportation systems. Firstly, we propose a near optimal ``mirror backpressure'' control methodology for joint entry/assignment/pricing control in a network where there are a fixed number of supply units (vehicles), and demands with different origin and destination nodes arrive over time. The MBP policy does not need demand arrival rate predictions at all, and we prove guarantees of near optimal performance over a finite horizon. Secondly, we study a special case of the network control problem where the geographical imbalances in demand are small enough such that, ignoring stochasticity, they can be corrected using assignment control alone. The objective is to minimize the fraction of customers who are ``lost'' (not served) because there is no vehicle at a nearby location when the customer arrives. We show that for this setting we can achieve a refined notion of optimality, i.e., the large deviations optimality. The second part of the dissertation analyzes equilibria in matching markets under different mechanisms. Firstly, we study the Gale-Shalpley ``deferred acceptance'' algorithm, which has been successfully adopted in contexts such as school choice and resident matching programs. Our research question is, ``Which Gale-Shapley matching markets exhibit a short-side advantage?'' I.e., in which markets does being on the short side of the market allow agents to obtain better match partners relative to a similar ``balanced'' market with equal numbers of agents on the two sides? We address this problem by looking at the ``random matching market'' model where each agent considers only a subset of potential partners on the other side, and sharply characterize the resulting (nearly unique) stable matching, overcoming significant technical challenges. Secondly, we study the waiting-list mechanism, which is commonly used in kidney assignment, public housing allocation, and beyond. We show that the waiting-list mechanism is near-optimal in terms of allocative efficiency for general systems with an arbitrary number of agent types and item types, and obtain tight bound on the efficiency loss. Comparing to existing works which could only analyze very simple systems, we tackle the general case by taking a completely different approach and establishing a novel connection with stochastic gradient descent.

This study examines the types of and frequency with which computer-mediated communications (CMC) services are used by international public relations (PR) specialists for advertising and information gathering and the judged effectiveness of these services. International PR specialists were surveyed to identify the types of and frequency with which CMC services were used for advertising and information gathering. Services focused on World Wide Web (V./WW) pages, electronic mail (e-mail), Internet search engines, and large commercial on-line services. This study also focused on: the use of the Internet for advertising, the judged effectiveness of using CMC for information gathering, the possible differences in the use of CMC between PR specialists from small large firms and a comparison of judged effectiveness based on firm size. A summary of CMC services was included to provide background information. The results indicated fairly high usage for information gathering and a lack of agreement regarding the effectiveness of CMC for advertising. No significant differences of type or frequency of usage were found when comparing small firms and large firms.

Internet Service Providers, or ISPs, like any other rational entity make decisions to maximize their profit. While some of their decisions are on how to attract customers, they inevitably need to control how much resources consumers utilize. In this dissertation, we focus on two different aspects of ISP's decisions, including bandwidth allocation and pricing techniques through which ISPs manage allotting their limited capacity to users with high demand, and zero-rating, which can be one of the tools through which the ISP can attract customers. For bandwidth allocation, this dissertation discusses the data plans available for each user's monthly billing cycle. Within those, the ISPs guarantee a fixed amount of data at high rates until a byte threshold is reached, at which point the user's data rate is throttled to a lower rate for the remainder of the cycle. In practice, the thresholds and rates of throttling can appear and may be somewhat arbitrary. In this dissertation, we evaluate the choice of threshold and rate as an optimization problem (regret minimization) and demonstrate that intuitive formulations of client regret, which preserve desirable fairness properties, lead to optimization problems that have tractably computable solutions. For zero-rating options in the ISP market, and their relation to net neutrality, we begin by introducing the concept of zero-rating, which refers to the practice of providing free Internet access to some users under certain conditions, and usually concurs with differentiation among users or content providers. Even though zero-rating is banned in some countries (India, Canada), others have either taken no stance or explicitly allowed it (South Africa, Kenya, U.S.). While there is broad agreement that preserving the content quality of service falls under the purview of net neutrality, the role of differential pricing, especially the practice of \emph{zero-rating} remains controversial. An objective of net neutrality is to design regulations for the Internet and ensure that it remains a public, open platform where innovations can thrive. We show the practice of zero-rating does not agree with that. This dissertation shows how ISPs could make zero-rating decisions to attract customers, and then show how these decisions may negatively impact the market and customer welfare, which necessitates the existence of some zero-rating regulations.

Submitted in partial fulfilment of the requirements for the degree of Masters in Business Administration (MBA), Business Studies Unit, Durban University of Technology, 2008 / The pharmaceutical industry is in the midst of a fundamental transformation. Time and cost constraints are forcing marketers to search for new ways to maintain and grow brand awareness. The amount of time that doctors allow for representatives to market their products is becoming less and less and as a result products are not getting the exposure they used to. Organisations that find innovative ways to maintain exposure of their products may gain a competitive advantage over those organisations that rely on traditional marketing methods. The prolific increase in the use of the internet may provide pharmaceutical organisations with a complementary channel to market their products. The objective of this study is to determine the levels of internet technology adoption among general practitioners. The Technology Acceptance Model (TAM) is one of the most influential research models in studies of the determinants of information systems/ information technology (IS/IT) acceptance. In TAM, perceived usefulness and perceived ease of use are hypothesised and empirically supported as fundamental determinants of user acceptance of a given IS/IT. Using TAM, this study sets out to determine general practitioners’ intention to use the internet as a source of pharmaceutical information. The literature review provides an extensive evaluation on the development of TAM and its application in different technologies. Based on these findings, the researcher developed this study to investigate internet technology acceptance in the pharmaceutical industry. Use is made of the descriptive survey method and data is retrieved from a sample of 105 general practitioners in Kwazulu-Natal. The observation is made via the benefit of a questionnaire. The process of sampling is that of convenient sampling. The analysis is quantitative and makes use of statistical analysis appropriate for the data. Analysis of the survey results produces useful insights into the factors influencing internet technology adoption by general practitioners. When analysing the independent variables, respondents were not in strong agreement about the perceived usefulness nor the perceived ease of use of internet technology as a source of pharmaceutical information. However, positive results from the relationships between the independent (perceived usefulness and perceived ease of use) and dependent variables (attitude and intention to use) provide valuable data with which organisations may develop internet-based marketing strategies. Based on the survey findings, recommendations using the Beynon-Davies (2004) Internet Adoption Model are suggested.

Submitted in partial fulfilment of the requirements for the degree of Masters in Business Administration (MBA), Business Studies Unit, Durban University of Technology, 2008 / The pharmaceutical industry is in the midst of a fundamental transformation. Time and cost constraints are forcing marketers to search for new ways to maintain and grow brand awareness. The amount of time that doctors allow for representatives to market their products is becoming less and less and as a result products are not getting the exposure they used to. Organisations that find innovative ways to maintain exposure of their products may gain a competitive advantage over those organisations that rely on traditional marketing methods. The prolific increase in the use of the internet may provide pharmaceutical organisations with a complementary channel to market their products. The objective of this study is to determine the levels of internet technology adoption among general practitioners. The Technology Acceptance Model (TAM) is one of the most influential research models in studies of the determinants of information systems/ information technology (IS/IT) acceptance. In TAM, perceived usefulness and perceived ease of use are hypothesised and empirically supported as fundamental determinants of user acceptance of a given IS/IT. Using TAM, this study sets out to determine general practitioners’ intention to use the internet as a source of pharmaceutical information. The literature review provides an extensive evaluation on the development of TAM and its application in different technologies. Based on these findings, the researcher developed this study to investigate internet technology acceptance in the pharmaceutical industry. Use is made of the descriptive survey method and data is retrieved from a sample of 105 general practitioners in Kwazulu-Natal. The observation is made via the benefit of a questionnaire. The process of sampling is that of convenient sampling. The analysis is quantitative and makes use of statistical analysis appropriate for the data. Analysis of the survey results produces useful insights into the factors influencing internet technology adoption by general practitioners. When analysing the independent variables, respondents were not in strong agreement about the perceived usefulness nor the perceived ease of use of internet technology as a source of pharmaceutical information. However, positive results from the relationships between the independent (perceived usefulness and perceived ease of use) and dependent variables (attitude and intention to use) provide valuable data with which organisations may develop internet-based marketing strategies. Based on the survey findings, recommendations using the Beynon-Davies (2004) Internet Adoption Model are suggested.

This study investigates the valuation of Internet-based businesses. In particular the influence of the implementation of an e-business strategy on the value of a business by focussing on its financial performance, will be determined. Although the valuation of businesses in general has been researched extensively, research on the valuation of Internet-based businesses produced contradictory findings. No consensus could be reached regarding the most appropriate valuation approach to be used. Some research findings indicated that the discounted cash flow approach was the most appropriate while others stipulated that a new valuation approach should be developed. Many authors state that the move to include an e-business strategy is natural, and that businesses cannot afford not to include some form of e-business strategy. Previous research has also shown that by including an e-business strategy, it is possible to improve efficiency of the business and ultimately increase profitability. However, there was no emphasis on how the e-business strategy will influence the business valuation. In order to establish whether an e-business strategy will create value for a business, an empirical investigation was undertaken.

Submitted in fulfillment of the requirements of the Master of Technology Degree in Information Technology, Durban University of Technology, Durban, South Africa, 2014. / Social computing researchers are devoting efforts to understand the complex social behaviour of people using social networking platforms, such as Twitter, LinkedIn and Facebook, so as to inform the design of human-centered and socially aware systems. This research study investigates the factors of perceived trust, user satisfaction, social norm and perceived behavioural control, to develop a model for predicting the continuance intention of people to use online social networking for business transactions. In order to validate the predictive capability of the model developed, an online survey was used to collect 300 useable responses from people who have used LinkedIn and Twitter social networking platforms for business transactions at least once. The Partial Least Square (PLS) mathematical analysis tool was thereafter used to perform confirmatory factor analysis, analysis of measurement and structural models. The study results provide significant evidence in support of the factors of perceived trust, social norm and user satisfaction, as determinants of the continuance intention of people using online social networking platforms for business transactions. Perceived trust was found to exhibit a strong relationship with social norm and explains a variance of (R2=0.47). In addition, social norm explains a variance of (R2=0.44) and user satisfaction explains a variance of (R2=0.42), resulting in the model predicting (R2=0.56) continuance intention. In addition, the research model was tested for the moderating effects of usage habit, which were found to significantly moderate relationships between continuance intention and perceived trust, PBCand social norm, resulting in an improved predictive capability of (R2=0.89). The moderating result indicates that a higher level of habit increases the effect of perceived trust, Perceived Behavioural Control (PBC) and social norm on continuance intention. This result confirms the theoretical argument that the strength of user satisfaction to predict continuance, is strengthened by usage habit. The results of this research study generally have practical implications for individuals who desire to offer commercial services on online social networking technologies, to seriously consider building trust and maintaining user satisfaction to sustain their businesses. They should also think of strategies embedded in peer pressure, to attract, retain and establish trustworthy relationships with customers.