This document analyzes Google Analytics data from an Iranian artist's website to understand factors that impact page views. It finds that direct visitors have the highest impact on page views, increasing them by 4.44 per visitor. Referral visitors also increase page views but only by 1.77 per visitor. Returning visitors similarly impact page views as direct visitors. Connection speed alone does not determine impact on views; the content and visitor location must also be considered.
An approach for evaluation of social...STIinnsbruck
This document describes an evaluation framework for social media monitoring tools. It proposes criteria to analyze these tools from three perspectives: concepts implemented, technologies used, and user interface. Concepts include analysis, insights, engagement, workflow management, and influence. Technologies include listening grid adjustment, near real-time processing, API integration, sentiment analysis, and access to historical data. The user interface is also important and should include a customizable dashboard and ability to export results. The evaluation framework is intended to help enterprises choose the right monitoring tool for their needs.
A novel approach to dynamic profiling of E-customers considering click stream...IJECEIAES
In this paper, we present an approach for mining change in customer’s behavior for the purpose of maintaining robust profiling model over time. Most of previous studies leave important questions unanswered: In developing B2C e-commerce strategies, how do managers implicitly load customer’s profiles based on their satisfaction over the online store characteristics? And: What kind of feedback segments do they have? Our proposed approach does not force customers to explicitly express their preference information over the online service but rather capture their preference from their online activities. The challenge does not only lay in analyzing how customer’s classifier model change and when it does so but also to adapt it to the customer’s click stream data using a new decision tree generation algorithm which takes as inputs new set of variables; categorical, continuous and fuzzy variables. Customer’s online reviews rates are considered as classes. Experiments show that this work performed well in identifying relevant customer’s stream data to judge the chinese e-commerce website “Tmall”. The extracted values of the website’s features are also useful to identifying the satisfaction level when the customer’s rate is not available.
The-State-Of-Internet-Marketing-Research-(2005-2012)--A-Systematic-Review-Usi...Janarthanan B
This study analyzed 2,158 research article abstracts from 460 journals between 2005 and 2012 to quantify internet marketing research trends. Consumer behavior was the most studied topic (21.87% of articles), followed by services (12.97%) and business strategy (12.37%). Purchase intention and social media were found to have high centrality among topics. The study provides an extensive review of internet marketing literature and insights into emerging research areas.
28th Workshop on Information Systems and EconomicsYunkun Zhao, PhD
The document discusses evaluating the effectiveness of different online customer touchpoints in omni-channel retailing environments. It identifies three gaps in previous research: a focus on distribution channels rather than information channels, consideration of only within-brand effects rather than cross-brand effects, and ignoring interdependencies among touchpoints. The study aims to address these gaps by analyzing a unique dataset from an omni-channel retailer to understand the relative impact of owned, paid, and earned media touchpoints on customer demand, including potential cross-brand effects and interdependencies among touchpoints. Preliminary results indicate customers on average spend 51 RMB, purchase 0.163 items with a 14.432% conversion rate given their exposures to different touchpoint media.
Measuring the quality of e-commerce websites using analytical hierarchy processTELKOMNIKA JOURNAL
Improving website quality for e-commerce website is indispensable since it affects customer satisfaction. There are several aspects of website quality that should be considered. Unfortunately, what criteria that should be prioritized is still under research. This research aims to identify the priority of website quality criteria and incorporate these criteria to measure the quality of ten e-commerce websites in Indonesia using Analytical Hierarchical Process (AHP). This study interviews two experts to assign priority for each criterion. In the end, this study has found that availability is the uttermost aspect to consider. Furthermore, this study also found that OLX.com is the best Customer-to-Customer (C2C) e-commerce in Indonesia in terms of website quality. This research is useful for any e-commerce technical developer to improve his/her website in twofold: 1) criteria priority to improve the quality of C2C e-commerce websites and 2) website quality ranking of C2C e-commerce.
Kano Model Servqual Customer Satisfaction Analysis in Retail BankingIJSRED
This document summarizes a research article that analyzes customer satisfaction in the retail banking industry in India using the Kano model. The Kano model is used to categorize banking attributes according to their impact on customer satisfaction. Attributes are prioritized so the bank can focus on attributes that provide the most customer delight. The study aims to provide an independent, third-party assessment of customer perceptions to help banks improve transparency and compliance.
E-commerce online review for detecting influencing factors users perceptionjournalBEEI
To date, online shopping using e-commerce services becomes a trend. The emergence of e-commerce truly helps people to shop more effectively and efficiently. However, there are still some problems encountered in e-commerce, especially from the user perspective. This research aims to explore user review data, particularly on factors that influence user perception of e-commerce applications, classify, and identify potential solutions to finding problems in e-commerce applications. Data is grabbed using web scraping techniques and classified using proper machine learning, i.e., support vector machine (SVM). Text associations and fishbone analysis are performed based on the classified user review data. The results of this study show that the user satisfaction problem can be captured. Furthermore, various services that should be provided as a potential solution to experienced customers' problems or application users' perception problems can be generated. A detailed discussion of these findings is available in this article.
THE IMPLEMENTATION OF MODEL SERVICE ORIENTED ARCHITECTURE TO ANALYZE STUDENT ...AM Publications
The application of Service Oriented Architecture (SOA) model may integrate one information system to others. Also, SOA can handle the complexity of hardware platforms, software, and other services development in the college. There are many services systems exist in the university, for instance student satisfaction system by questionnaire. Student satisfaction is very important to be used as an evaluation of the college. Student satisfaction was analyzed using Importance Performance Analysis (IPA) by using quadrants for each competency. This study discusses the application of SOA and IPA model to analyze student satisfaction towards the course. Student satisfaction data was taken using questionnaires sent to the student satisfaction information system for 70 respondents for each course. It is found that sending the data using SOA and JavaScript Object Notation (JSON) is very efficient with the average value of page generation time for 1,5414 seconds. In addition, the analysis of student satisfaction towards each course is ranged from 3,72 to 4,09 with a satisfaction scale of 1 - 5. Therefore, the level of student satisfaction at the course is categorized in the satisfactory scale.
An approach for evaluation of social...STIinnsbruck
This document describes an evaluation framework for social media monitoring tools. It proposes criteria to analyze these tools from three perspectives: concepts implemented, technologies used, and user interface. Concepts include analysis, insights, engagement, workflow management, and influence. Technologies include listening grid adjustment, near real-time processing, API integration, sentiment analysis, and access to historical data. The user interface is also important and should include a customizable dashboard and ability to export results. The evaluation framework is intended to help enterprises choose the right monitoring tool for their needs.
A novel approach to dynamic profiling of E-customers considering click stream...IJECEIAES
In this paper, we present an approach for mining change in customer’s behavior for the purpose of maintaining robust profiling model over time. Most of previous studies leave important questions unanswered: In developing B2C e-commerce strategies, how do managers implicitly load customer’s profiles based on their satisfaction over the online store characteristics? And: What kind of feedback segments do they have? Our proposed approach does not force customers to explicitly express their preference information over the online service but rather capture their preference from their online activities. The challenge does not only lay in analyzing how customer’s classifier model change and when it does so but also to adapt it to the customer’s click stream data using a new decision tree generation algorithm which takes as inputs new set of variables; categorical, continuous and fuzzy variables. Customer’s online reviews rates are considered as classes. Experiments show that this work performed well in identifying relevant customer’s stream data to judge the chinese e-commerce website “Tmall”. The extracted values of the website’s features are also useful to identifying the satisfaction level when the customer’s rate is not available.
The-State-Of-Internet-Marketing-Research-(2005-2012)--A-Systematic-Review-Usi...Janarthanan B
This study analyzed 2,158 research article abstracts from 460 journals between 2005 and 2012 to quantify internet marketing research trends. Consumer behavior was the most studied topic (21.87% of articles), followed by services (12.97%) and business strategy (12.37%). Purchase intention and social media were found to have high centrality among topics. The study provides an extensive review of internet marketing literature and insights into emerging research areas.
28th Workshop on Information Systems and EconomicsYunkun Zhao, PhD
The document discusses evaluating the effectiveness of different online customer touchpoints in omni-channel retailing environments. It identifies three gaps in previous research: a focus on distribution channels rather than information channels, consideration of only within-brand effects rather than cross-brand effects, and ignoring interdependencies among touchpoints. The study aims to address these gaps by analyzing a unique dataset from an omni-channel retailer to understand the relative impact of owned, paid, and earned media touchpoints on customer demand, including potential cross-brand effects and interdependencies among touchpoints. Preliminary results indicate customers on average spend 51 RMB, purchase 0.163 items with a 14.432% conversion rate given their exposures to different touchpoint media.
Measuring the quality of e-commerce websites using analytical hierarchy processTELKOMNIKA JOURNAL
Improving website quality for e-commerce website is indispensable since it affects customer satisfaction. There are several aspects of website quality that should be considered. Unfortunately, what criteria that should be prioritized is still under research. This research aims to identify the priority of website quality criteria and incorporate these criteria to measure the quality of ten e-commerce websites in Indonesia using Analytical Hierarchical Process (AHP). This study interviews two experts to assign priority for each criterion. In the end, this study has found that availability is the uttermost aspect to consider. Furthermore, this study also found that OLX.com is the best Customer-to-Customer (C2C) e-commerce in Indonesia in terms of website quality. This research is useful for any e-commerce technical developer to improve his/her website in twofold: 1) criteria priority to improve the quality of C2C e-commerce websites and 2) website quality ranking of C2C e-commerce.
Kano Model Servqual Customer Satisfaction Analysis in Retail BankingIJSRED
This document summarizes a research article that analyzes customer satisfaction in the retail banking industry in India using the Kano model. The Kano model is used to categorize banking attributes according to their impact on customer satisfaction. Attributes are prioritized so the bank can focus on attributes that provide the most customer delight. The study aims to provide an independent, third-party assessment of customer perceptions to help banks improve transparency and compliance.
E-commerce online review for detecting influencing factors users perceptionjournalBEEI
To date, online shopping using e-commerce services becomes a trend. The emergence of e-commerce truly helps people to shop more effectively and efficiently. However, there are still some problems encountered in e-commerce, especially from the user perspective. This research aims to explore user review data, particularly on factors that influence user perception of e-commerce applications, classify, and identify potential solutions to finding problems in e-commerce applications. Data is grabbed using web scraping techniques and classified using proper machine learning, i.e., support vector machine (SVM). Text associations and fishbone analysis are performed based on the classified user review data. The results of this study show that the user satisfaction problem can be captured. Furthermore, various services that should be provided as a potential solution to experienced customers' problems or application users' perception problems can be generated. A detailed discussion of these findings is available in this article.
THE IMPLEMENTATION OF MODEL SERVICE ORIENTED ARCHITECTURE TO ANALYZE STUDENT ...AM Publications
The application of Service Oriented Architecture (SOA) model may integrate one information system to others. Also, SOA can handle the complexity of hardware platforms, software, and other services development in the college. There are many services systems exist in the university, for instance student satisfaction system by questionnaire. Student satisfaction is very important to be used as an evaluation of the college. Student satisfaction was analyzed using Importance Performance Analysis (IPA) by using quadrants for each competency. This study discusses the application of SOA and IPA model to analyze student satisfaction towards the course. Student satisfaction data was taken using questionnaires sent to the student satisfaction information system for 70 respondents for each course. It is found that sending the data using SOA and JavaScript Object Notation (JSON) is very efficient with the average value of page generation time for 1,5414 seconds. In addition, the analysis of student satisfaction towards each course is ranged from 3,72 to 4,09 with a satisfaction scale of 1 - 5. Therefore, the level of student satisfaction at the course is categorized in the satisfactory scale.
IRJET- Web Traffic Analysis through Data Analysis and Machine LearningIRJET Journal
This document discusses analyzing web traffic through data analysis and machine learning. It begins by introducing the need for web traffic analysis as internet usage increases. It then discusses collecting and processing web usage data, key metrics for analysis like number of users and page views, and identifying important performance indicators. The document outlines different methods of analyzing web traffic, both online and offline. It discusses using machine learning to accept web traffic data, clean it, and analyze patterns. The conclusion reiterates the importance of monitoring web traffic to provide sufficient capacity for users.
1) The author conducted a website audit and analysis of the Region of Waterloo International Airport website to improve its effectiveness as a marketing tool.
2) Key aspects of the audit included analyzing traffic, bounce rates, and content on each page to determine needed changes. Outdated content, lack of internal links, and missing image descriptions were identified.
3) Based on the audit, the author implemented search engine optimization strategies like adding internal links between pages, rewriting and updating content, and including image descriptions. This was done to better retain visitors and attract new users to the airport website.
This document provides an introduction and background for a thesis on internet marketing acquisition processes. It discusses internet marketing and how it relates to the traditional marketing mix. It also explores companies' and consumers' perspectives on internet marketing. The thesis will focus on analyzing the acquisition process for attracting visitors to the website www.flashgame4fun.com, specifically looking at search engine optimization and link building activities. Both qualitative and quantitative research methods will be used, including analyzing documents, collecting data from the case study website, and testing the effectiveness of acquisition activities.
A Fuzzy Expert System for Assessing the InternetStores Based on Web Site Attr...IDES Editor
Purpose of this research is determining the success
of online shopping stores by an intelligent system. Here a
Fuzzy Expert System has been designed with the consideration
of website attributes as input variables. The web site success
level is determined in this system as output.
The rules of systems have been extracted from some ecommerce
experts and the systems have been developed with
the use of FIS tool of MATLAB software. The final result
contains an anticipating model for evaluating level of web
site success of shopping centers based on website factors
situation. The presented steps have been run in four online
bookstores as the empirical study.
WEB BASED INFORMATION SYSTEMS OF E-COMMERCE USER SATISFACTION USING ZACHMAN ...AM Publications
The progress in information and communication technologies and the birth of the internet has changed the end user computing experience and environment. This advancement has changed the way of delivering information and services. Information systems can now be web-enabled, unlike the traditional stationary information systems. Web based means the information system includes in the website. Web portals are a part of this advancement. A Web portal is a gateway to information and services from multiple sources in a unified way, using a single, unique user interface. This study will make design of web base online information systems of e-commerce user satisfaction using Zachman framework. The major objectives of this study was to investigate how information system IS measures that influence user satisfaction of the Diponegoro university students, also to design and analysis the implementation of Zachman Framework. Based on the data analysis collected by e-mail, the conclusion were: Design of the website in online store is good implementation because there is majority of respondents agree for customer satisfaction. The analysis of zachman framework for shopme design online store reflected the best solution because describe the more comprehensive coverage for all enterprise architecture stakeholders. The information system is measures the influence user satisfaction of the Diponegoro University students, and the design and analysis the implementation of zachman framework in this study presented a method that provides guidance in the development of an organization’s enterprise architecture.
RESEARCH CHALLENGES IN WEB ANALYTICS – A STUDYIRJET Journal
This document discusses research challenges in web analytics. It begins with an introduction to web analytics and its uses and features such as improving user experience, enhancing e-commerce, and tracking marketing campaigns. Some common web analytics tools are then described such as Google Analytics, Excel, and Tableau. The web analytics process involves defining goals, collecting and analyzing data, reporting results, developing online strategies, and testing improvements. Key research challenges discussed include data privacy issues, evaluating e-commerce sites using tools like Google Analytics, and using analytics to improve access to archival resources online.
Product Comparison Website using Web scraping and Machine learning.IRJET Journal
This document describes the development of a product comparison website using web scraping and machine learning techniques. The website extracts data like price, features and ratings from multiple e-commerce websites using web scraping. A machine learning algorithm then compares the collected data and provides personalized recommendations to users. The implementation demonstrated that web scraping is effective for collecting product data for comparison. Future work may involve incorporating deep learning models and external data sources to improve recommendations.
From an experimental ISTAT survey:
1. Researchers used web scraping and text mining of 85,000 Italian company websites to develop models predicting key metrics like web ordering.
2. They compared these "alternative estimates" to traditional survey estimates for accuracy, finding the new estimates were equally accurate.
3. The new methods provided additional granular data like estimates by industry and region not available from surveys alone.
This document discusses a study analyzing the relationships between four areas of technology-enabled retail services (content management, customer management, multi-channel management, and visitor traffic management) and the online sales performance of web retailers. The study uses data on the profiles, operations, and sales of the top 500 US web retailers to examine how their implementation of features in these four areas impacts online sales. The results show that content, customer, and multi-channel management features have a significant positive impact on sales, while the impact of visitor traffic management features is inconclusive.
Dimensions of business to consumer (b2 c) systems success in kuwaitsan18
This document summarizes an academic journal article that tests an enhanced model of business-to-consumer (B2C) e-commerce systems success in Kuwait. The study builds on prior research by Wang (2008) that applied the Delone and McLean information systems success model to e-commerce. The proposed model adds several improvements, including a multidimensional view of system quality, conceptualizations of both monetary and non-monetary value, and e-commerce-specific factors. The model was tested using 288 experienced B2C consumers in Kuwait and the results largely support the hypothesized relationships in the model. A key finding is that while service quality influences perceived value, it does not impact user satisfaction in the Arab context,
A usability evaluation framework for b2 c e commerce websitesAlexander Decker
This document presents a framework for evaluating the usability of B2C e-commerce websites. It involves user testing methods like usability testing and interviews to identify usability problems in areas like navigation, design, purchasing processes, and customer service. The framework specifies goals for the evaluation, determines which website aspects to evaluate, and identifies target users. It then describes collecting data through user testing and analyzing the results to identify usability problems and suggest improvements.
A usability evaluation framework for b2 c e commerce websitesAlexander Decker
This document presents a framework for evaluating the usability of B2C e-commerce websites. It involves user testing methods like usability testing and interviews to identify usability problems in areas like navigation, design, purchasing processes, and customer service. The framework specifies goals for the evaluation, determines which website aspects to evaluate, and identifies target users. It then describes collecting data through user testing and analyzing the results to identify usability problems and suggest improvements.
IRJET- Content Analysis of Websites of Health Ministries in ECOWAS English Sp...IRJET Journal
This document evaluates the websites of the Ministries of Health of five English-speaking countries in the Economic Community of West African States (ECOWAS) using the Website Attribute Evaluation System (WAES). The evaluation found that the websites were at a similar basic level of development, focusing primarily on information dissemination. Content analysis was conducted on attributes such as information provided, page layout, multimedia elements, and rate of updates. The results showed minor variations across countries but overall consistency in the basic level of development of the health ministry websites.
This document discusses predicting and optimizing e-commerce conversion rates through supervised and unsupervised machine learning algorithms. It first reviews literature on predicting conversions through clickstream data analysis and machine learning approaches. It then describes the objectives of this study, which are to determine key factors that drive conversion and produce an accurate predictive model. The methods section outlines a three stage approach: 1) Analyzing conversion rates by marketing channel 2) Clustering ads within a key channel by performance 3) Predictive modeling using multiple variables. Key performance ratios, hierarchical clustering, and several machine regression algorithms are implemented and compared.
The document discusses key performance indicators (KPIs) and metrics for measuring digital marketing effectiveness. It conducted a systematic literature review of 378 sources on digital marketing and web analytics to identify the most relevant KPIs. The review found that common KPIs included search engine optimization metrics like click-through rate, as well as metrics for paid search, social media, traffic, leads, conversions and ROI. The study aims to help marketers understand which metrics are most important for increasing the effectiveness of digital marketing strategies.
Smart Engine analytics allows you to analyze large volumes of data over extended periods of time, utilizing algorithms to determine precise data from Web traffic and feeding this data to the reporting components. With its industry-leading algorithms, the key benefit of Wavecrest’s Smart Engine is providing the manager-ready, actionable reporting that nontechnical personnel in the company needs.
Requirement analysis method of e commerce websites development for small-medi...ijseajournal
Along with the growth of the Internet, the trend shows that e-commerce have been growing significantly in
the last several years. This means business opportunities for small-medium enterprises (SMEs), which are
recognized as the backbone of the economy. SMEs may develop and run small to medium size of particular
e-commerce websites as the solution of specific business opportunities. Certainly, the websites should be
developed accordingly to support business success. In developing the websites, key elements of e-commerce
business model that are necessary to ensure the success should be resolved at the requirement stage of the
development. In this paper, we propose an enhancement of requirement analysis method found in
literatures such that it includes activities to resolve the key elements. The method has been applied in three
case studies based on Indonesia situations and we conclude that it is suitable to be adopted by SMEs.
This document describes a website health checker system that monitors websites and sends alert messages if issues are detected. It discusses the need for website monitoring to ensure high performance and availability. The proposed system uses a cron job to continuously send ICMP requests to monitored websites and triggers alerts via email or SMS if response times exceed thresholds. It is implemented using the Flask web framework with a user interface to add domains and view monitoring results. The system aims to rapidly detect and correct problems before users are impacted through real-time alerts.
E-Commerce Mobile Marketing Model Resolving Users Acceptance Criteria IJMIT JOURNAL
The growth of e-commerce and mobile services has been creating business opportunities. Among these,
mobile marketing is predicted to be a new trend of marketing. As mobile devices are personal tools such
that the services ought to be unique, in the last few years, researches have been conducted studies related
to the user acceptance of mobile services and produced results. This research aims to develop a model of
mobile e-commerce mobile marketing system, in the form of computer-based information system (CBIS)
that addresses the recommendations or criteria formulated by researchers. In this paper, the criteria
formulated by researches are presented then each of the criteria is resolved and translated into mobile
services. A model of CBIS, which is an integration of a website and a mobile application, is designed to
materialize the mobile services. The model is presented in the form of business model, system procedures,
network topology, software model of the website and mobile application, and database models.
Data Mining to Facilitate Effective User Navigation and Improve Structure of ...IJERA Editor
Now a days for a company it is very important to have active presence on web to become successful in electronic market. This requirement is fulfilled by an interactive website of a company. Website can be used to sell goods, maintain customer relationships, promotion of goods. By looking at the customer’s response towards website we can decide campaign for future products and services. So for this website should be interactive and must be used by user by the way designer wants to. So, it is very important to study navigational behavior of a customer. By analyzing the web logs which records the navigational activity of the customer we can get several sequential patterns which helps us to study whether user is pursuing site’s goal or not. Sequences are analyzed with WUM (Web Utilization Miner) which gives g-sequence (Generalized Sequence) and aggregate tree as an output.
IRJET- Web Traffic Analysis through Data Analysis and Machine LearningIRJET Journal
This document discusses analyzing web traffic through data analysis and machine learning. It begins by introducing the need for web traffic analysis as internet usage increases. It then discusses collecting and processing web usage data, key metrics for analysis like number of users and page views, and identifying important performance indicators. The document outlines different methods of analyzing web traffic, both online and offline. It discusses using machine learning to accept web traffic data, clean it, and analyze patterns. The conclusion reiterates the importance of monitoring web traffic to provide sufficient capacity for users.
1) The author conducted a website audit and analysis of the Region of Waterloo International Airport website to improve its effectiveness as a marketing tool.
2) Key aspects of the audit included analyzing traffic, bounce rates, and content on each page to determine needed changes. Outdated content, lack of internal links, and missing image descriptions were identified.
3) Based on the audit, the author implemented search engine optimization strategies like adding internal links between pages, rewriting and updating content, and including image descriptions. This was done to better retain visitors and attract new users to the airport website.
This document provides an introduction and background for a thesis on internet marketing acquisition processes. It discusses internet marketing and how it relates to the traditional marketing mix. It also explores companies' and consumers' perspectives on internet marketing. The thesis will focus on analyzing the acquisition process for attracting visitors to the website www.flashgame4fun.com, specifically looking at search engine optimization and link building activities. Both qualitative and quantitative research methods will be used, including analyzing documents, collecting data from the case study website, and testing the effectiveness of acquisition activities.
A Fuzzy Expert System for Assessing the InternetStores Based on Web Site Attr...IDES Editor
Purpose of this research is determining the success
of online shopping stores by an intelligent system. Here a
Fuzzy Expert System has been designed with the consideration
of website attributes as input variables. The web site success
level is determined in this system as output.
The rules of systems have been extracted from some ecommerce
experts and the systems have been developed with
the use of FIS tool of MATLAB software. The final result
contains an anticipating model for evaluating level of web
site success of shopping centers based on website factors
situation. The presented steps have been run in four online
bookstores as the empirical study.
WEB BASED INFORMATION SYSTEMS OF E-COMMERCE USER SATISFACTION USING ZACHMAN ...AM Publications
The progress in information and communication technologies and the birth of the internet has changed the end user computing experience and environment. This advancement has changed the way of delivering information and services. Information systems can now be web-enabled, unlike the traditional stationary information systems. Web based means the information system includes in the website. Web portals are a part of this advancement. A Web portal is a gateway to information and services from multiple sources in a unified way, using a single, unique user interface. This study will make design of web base online information systems of e-commerce user satisfaction using Zachman framework. The major objectives of this study was to investigate how information system IS measures that influence user satisfaction of the Diponegoro university students, also to design and analysis the implementation of Zachman Framework. Based on the data analysis collected by e-mail, the conclusion were: Design of the website in online store is good implementation because there is majority of respondents agree for customer satisfaction. The analysis of zachman framework for shopme design online store reflected the best solution because describe the more comprehensive coverage for all enterprise architecture stakeholders. The information system is measures the influence user satisfaction of the Diponegoro University students, and the design and analysis the implementation of zachman framework in this study presented a method that provides guidance in the development of an organization’s enterprise architecture.
RESEARCH CHALLENGES IN WEB ANALYTICS – A STUDYIRJET Journal
This document discusses research challenges in web analytics. It begins with an introduction to web analytics and its uses and features such as improving user experience, enhancing e-commerce, and tracking marketing campaigns. Some common web analytics tools are then described such as Google Analytics, Excel, and Tableau. The web analytics process involves defining goals, collecting and analyzing data, reporting results, developing online strategies, and testing improvements. Key research challenges discussed include data privacy issues, evaluating e-commerce sites using tools like Google Analytics, and using analytics to improve access to archival resources online.
Product Comparison Website using Web scraping and Machine learning.IRJET Journal
This document describes the development of a product comparison website using web scraping and machine learning techniques. The website extracts data like price, features and ratings from multiple e-commerce websites using web scraping. A machine learning algorithm then compares the collected data and provides personalized recommendations to users. The implementation demonstrated that web scraping is effective for collecting product data for comparison. Future work may involve incorporating deep learning models and external data sources to improve recommendations.
From an experimental ISTAT survey:
1. Researchers used web scraping and text mining of 85,000 Italian company websites to develop models predicting key metrics like web ordering.
2. They compared these "alternative estimates" to traditional survey estimates for accuracy, finding the new estimates were equally accurate.
3. The new methods provided additional granular data like estimates by industry and region not available from surveys alone.
This document discusses a study analyzing the relationships between four areas of technology-enabled retail services (content management, customer management, multi-channel management, and visitor traffic management) and the online sales performance of web retailers. The study uses data on the profiles, operations, and sales of the top 500 US web retailers to examine how their implementation of features in these four areas impacts online sales. The results show that content, customer, and multi-channel management features have a significant positive impact on sales, while the impact of visitor traffic management features is inconclusive.
Dimensions of business to consumer (b2 c) systems success in kuwaitsan18
This document summarizes an academic journal article that tests an enhanced model of business-to-consumer (B2C) e-commerce systems success in Kuwait. The study builds on prior research by Wang (2008) that applied the Delone and McLean information systems success model to e-commerce. The proposed model adds several improvements, including a multidimensional view of system quality, conceptualizations of both monetary and non-monetary value, and e-commerce-specific factors. The model was tested using 288 experienced B2C consumers in Kuwait and the results largely support the hypothesized relationships in the model. A key finding is that while service quality influences perceived value, it does not impact user satisfaction in the Arab context,
A usability evaluation framework for b2 c e commerce websitesAlexander Decker
This document presents a framework for evaluating the usability of B2C e-commerce websites. It involves user testing methods like usability testing and interviews to identify usability problems in areas like navigation, design, purchasing processes, and customer service. The framework specifies goals for the evaluation, determines which website aspects to evaluate, and identifies target users. It then describes collecting data through user testing and analyzing the results to identify usability problems and suggest improvements.
A usability evaluation framework for b2 c e commerce websitesAlexander Decker
This document presents a framework for evaluating the usability of B2C e-commerce websites. It involves user testing methods like usability testing and interviews to identify usability problems in areas like navigation, design, purchasing processes, and customer service. The framework specifies goals for the evaluation, determines which website aspects to evaluate, and identifies target users. It then describes collecting data through user testing and analyzing the results to identify usability problems and suggest improvements.
IRJET- Content Analysis of Websites of Health Ministries in ECOWAS English Sp...IRJET Journal
This document evaluates the websites of the Ministries of Health of five English-speaking countries in the Economic Community of West African States (ECOWAS) using the Website Attribute Evaluation System (WAES). The evaluation found that the websites were at a similar basic level of development, focusing primarily on information dissemination. Content analysis was conducted on attributes such as information provided, page layout, multimedia elements, and rate of updates. The results showed minor variations across countries but overall consistency in the basic level of development of the health ministry websites.
This document discusses predicting and optimizing e-commerce conversion rates through supervised and unsupervised machine learning algorithms. It first reviews literature on predicting conversions through clickstream data analysis and machine learning approaches. It then describes the objectives of this study, which are to determine key factors that drive conversion and produce an accurate predictive model. The methods section outlines a three stage approach: 1) Analyzing conversion rates by marketing channel 2) Clustering ads within a key channel by performance 3) Predictive modeling using multiple variables. Key performance ratios, hierarchical clustering, and several machine regression algorithms are implemented and compared.
The document discusses key performance indicators (KPIs) and metrics for measuring digital marketing effectiveness. It conducted a systematic literature review of 378 sources on digital marketing and web analytics to identify the most relevant KPIs. The review found that common KPIs included search engine optimization metrics like click-through rate, as well as metrics for paid search, social media, traffic, leads, conversions and ROI. The study aims to help marketers understand which metrics are most important for increasing the effectiveness of digital marketing strategies.
Smart Engine analytics allows you to analyze large volumes of data over extended periods of time, utilizing algorithms to determine precise data from Web traffic and feeding this data to the reporting components. With its industry-leading algorithms, the key benefit of Wavecrest’s Smart Engine is providing the manager-ready, actionable reporting that nontechnical personnel in the company needs.
Requirement analysis method of e commerce websites development for small-medi...ijseajournal
Along with the growth of the Internet, the trend shows that e-commerce have been growing significantly in
the last several years. This means business opportunities for small-medium enterprises (SMEs), which are
recognized as the backbone of the economy. SMEs may develop and run small to medium size of particular
e-commerce websites as the solution of specific business opportunities. Certainly, the websites should be
developed accordingly to support business success. In developing the websites, key elements of e-commerce
business model that are necessary to ensure the success should be resolved at the requirement stage of the
development. In this paper, we propose an enhancement of requirement analysis method found in
literatures such that it includes activities to resolve the key elements. The method has been applied in three
case studies based on Indonesia situations and we conclude that it is suitable to be adopted by SMEs.
This document describes a website health checker system that monitors websites and sends alert messages if issues are detected. It discusses the need for website monitoring to ensure high performance and availability. The proposed system uses a cron job to continuously send ICMP requests to monitored websites and triggers alerts via email or SMS if response times exceed thresholds. It is implemented using the Flask web framework with a user interface to add domains and view monitoring results. The system aims to rapidly detect and correct problems before users are impacted through real-time alerts.
E-Commerce Mobile Marketing Model Resolving Users Acceptance Criteria IJMIT JOURNAL
The growth of e-commerce and mobile services has been creating business opportunities. Among these,
mobile marketing is predicted to be a new trend of marketing. As mobile devices are personal tools such
that the services ought to be unique, in the last few years, researches have been conducted studies related
to the user acceptance of mobile services and produced results. This research aims to develop a model of
mobile e-commerce mobile marketing system, in the form of computer-based information system (CBIS)
that addresses the recommendations or criteria formulated by researchers. In this paper, the criteria
formulated by researches are presented then each of the criteria is resolved and translated into mobile
services. A model of CBIS, which is an integration of a website and a mobile application, is designed to
materialize the mobile services. The model is presented in the form of business model, system procedures,
network topology, software model of the website and mobile application, and database models.
Data Mining to Facilitate Effective User Navigation and Improve Structure of ...IJERA Editor
Now a days for a company it is very important to have active presence on web to become successful in electronic market. This requirement is fulfilled by an interactive website of a company. Website can be used to sell goods, maintain customer relationships, promotion of goods. By looking at the customer’s response towards website we can decide campaign for future products and services. So for this website should be interactive and must be used by user by the way designer wants to. So, it is very important to study navigational behavior of a customer. By analyzing the web logs which records the navigational activity of the customer we can get several sequential patterns which helps us to study whether user is pursuing site’s goal or not. Sequences are analyzed with WUM (Web Utilization Miner) which gives g-sequence (Generalized Sequence) and aggregate tree as an output.
Similar to Time Series ANN Approach for Weather Forecasting (20)
Temple of Asclepius in Thrace. Excavation resultsKrassimira Luka
The temple and the sanctuary around were dedicated to Asklepios Zmidrenus. This name has been known since 1875 when an inscription dedicated to him was discovered in Rome. The inscription is dated in 227 AD and was left by soldiers originating from the city of Philippopolis (modern Plovdiv).
CapTechTalks Webinar Slides June 2024 Donovan Wright.pptxCapitolTechU
Slides from a Capitol Technology University webinar held June 20, 2024. The webinar featured Dr. Donovan Wright, presenting on the Department of Defense Digital Transformation.
Gender and Mental Health - Counselling and Family Therapy Applications and In...PsychoTech Services
A proprietary approach developed by bringing together the best of learning theories from Psychology, design principles from the world of visualization, and pedagogical methods from over a decade of training experience, that enables you to: Learn better, faster!
A Visual Guide to 1 Samuel | A Tale of Two HeartsSteve Thomason
These slides walk through the story of 1 Samuel. Samuel is the last judge of Israel. The people reject God and want a king. Saul is anointed as the first king, but he is not a good king. David, the shepherd boy is anointed and Saul is envious of him. David shows honor while Saul continues to self destruct.
A Free 200-Page eBook ~ Brain and Mind Exercise.pptxOH TEIK BIN
(A Free eBook comprising 3 Sets of Presentation of a selection of Puzzles, Brain Teasers and Thinking Problems to exercise both the mind and the Right and Left Brain. To help keep the mind and brain fit and healthy. Good for both the young and old alike.
Answers are given for all the puzzles and problems.)
With Metta,
Bro. Oh Teik Bin 🙏🤓🤔🥰
🔥🔥🔥🔥🔥🔥🔥🔥🔥
إضغ بين إيديكم من أقوى الملازم التي صممتها
ملزمة تشريح الجهاز الهيكلي (نظري 3)
💀💀💀💀💀💀💀💀💀💀
تتميز هذهِ الملزمة بعِدة مُميزات :
1- مُترجمة ترجمة تُناسب جميع المستويات
2- تحتوي على 78 رسم توضيحي لكل كلمة موجودة بالملزمة (لكل كلمة !!!!)
#فهم_ماكو_درخ
3- دقة الكتابة والصور عالية جداً جداً جداً
4- هُنالك بعض المعلومات تم توضيحها بشكل تفصيلي جداً (تُعتبر لدى الطالب أو الطالبة بإنها معلومات مُبهمة ومع ذلك تم توضيح هذهِ المعلومات المُبهمة بشكل تفصيلي جداً
5- الملزمة تشرح نفسها ب نفسها بس تكلك تعال اقراني
6- تحتوي الملزمة في اول سلايد على خارطة تتضمن جميع تفرُعات معلومات الجهاز الهيكلي المذكورة في هذهِ الملزمة
واخيراً هذهِ الملزمة حلالٌ عليكم وإتمنى منكم إن تدعولي بالخير والصحة والعافية فقط
كل التوفيق زملائي وزميلاتي ، زميلكم محمد الذهبي 💊💊
🔥🔥🔥🔥🔥🔥🔥🔥🔥
Elevate Your Nonprofit's Online Presence_ A Guide to Effective SEO Strategies...TechSoup
Whether you're new to SEO or looking to refine your existing strategies, this webinar will provide you with actionable insights and practical tips to elevate your nonprofit's online presence.
Elevate Your Nonprofit's Online Presence_ A Guide to Effective SEO Strategies...
Time Series ANN Approach for Weather Forecasting
1. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
DOI : 10.5121/ijwest.2011.2102 14
ANALYZING THE IMPACT OF VISITORS ON PAGE
VIEWS WITH GOOGLE ANALYTICS
MOHAMMAD AMIN OMIDVAR
1
, VAHID REZA MIRABI
2
AND NARJES SHOKRY
3
1
Faculty of Information Technology Management, Iran Islamic Azad University E-
campus
Amin.omidvar@gmail.com
2
Faculty of Management, Islamic Azad University Central branch (Tehran), Iran
vrmirabi@yahoo.com
3
Faculty of Public Administration, Islamic Azad University Central branch (Tehran),
Iran
nargesshokry@yahoo.com
ABSTRACT
This paper develops a flexible methodology to analyze the effectiveness of different variables on
various dependent variables which all are times series and especially shows how to use a time
series regression on one of the most important and primary index (page views per visit) on
Google analytic and in conjunction it shows how to use the most suitable data to gain a more
accurate result.
Search engine visitors have a variety of impact on page views which cannot be described by
single regression. On one hand referral visitors are well-fitted on linear regression with low
impact. On the other hand, direct visitors made a huge impact on page views. The higher
connection speed does not simply imply higher impact on page views and the content of web
page and the territory of visitors can help connection speed to describe user behavior.
Returning visitors have some similarities with direct visitors.
KEYWORDS
Internet, User studies, worldwide web, Systems analysis, Data mining, visitors behavior, web
analysis, web metric, Google Analytics
1. Introduction
The internet is growingly rapidly and has a great impact on many businesses. Thousands of
companies now own a website and websites have become an integrated part of the business.
Furthermore many companies have employed many technologies which are available through
the web such as online services. With web information, web developers and designers can
improve user interfaces, search engines, navigation features, online help and information
architecture and have happier visitors/ costumers [21]. One of the most popular ways which
most frequented websites use to collect data and information about their websites is through
web analytic. Web analytic collects a large amount of data from users such as browser type,
connection speed, screen size, visitors’ type, etc. The collected data are usually large in
quantity and type that need to be further processed to become useful information or knowledge.
2. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
15
2. Literature review
The internet has been playing the important role of corporate marketing during the past ten
years [43]. With its combination of rich text, multimedia and user involvement, the internet
contains more information than any other media [26], [38]. The internet offers speed, reach, and
multimedia advantages, and has changed the way in which businesses interact with their
customers, suppliers, competitors, and employees [12].
Nearly all businesses now have a website [17]. A corporate website enriches the image of a
business and provides direct benefits in terms of electronic commerce (e-commerce) sales [30]
and indirect benefits in terms of information retrieval, branding, and services [34]. Recognition
of the internet is driving marketers in traditional companies to conduct transactions on the
internet [15]. Barua, Konana, B.Whinston, & yin (2001) found that e-business operational
excellence results in financial performance [8]. Thousands of companies have a fear to be left
behind by their competitors if they do not use online technologies.
The total number of internet users and the number of websites are increasing significantly
(table 1-1, table 1-2) which will result in the rapid growth of the use of the world wide web for
commercial purposes.
Figure 1, (Global Number of Internet Users,total and per 100 inhabitants 2000-2009)
3. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
16
Figure 2 Internet Domain Survey Host Count(Consotium, 2010)
On a five year forecast by Forrester, e-commerce sales in the U.S. will grow 10% annually from
2014 and online retail sales will be nearly $250 billion, up from $155 billion in 2009 [40].
The increasing amount of internet users, websites and retail sales implies that web developing
should be carried out in a competent, professional manner to increase profit. However,
systematic analysis of costumers’ behaviors has not kept pace with the rapid growth in e-
commerce. Without quantifiable metrics which is available through web analytics software,
website optimization (WSO) is a guessing game, therefore a majority of e-commerce companies
cannot afford this risk given their huge amount of money. Above 70% of the most frequented
websites use web analytic tools but with their large amount of data, it is difficult to use them
effectively[1]. Therefore, it is important to understand what kind of data and knowledge are
required for successful website development work.
Web Analytics Association Standards (2006) committee defined the three most important
metrics as Unique Visitors, Visits/Sessions, and Page Views; and, also categorized search
engine marketing metrics through counts (visits…), ratios (page views per visits….), and key
performance indicators (KPIs) [5].
The main reasons for measuring Search engine marketing (SEM) successes are related to traffic
measurement and the return on investment come on 4th [41]. The top for reasons are as follows:
• Increased traffic volume (76%)
• Conversion rates (76%)
• Click-through rates or CTRs (70%)
• Return on investment (67%)
4. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
17
Figure 3 Metrics used to measure Search Engine Marketing campaign success
Progressive improvement of SEM campaigns, conversion rates, and website performance are
available through web metrics, which would results in an increase in profits, happier customers,
and higher return on investment (ROI) by tracking progress over time or against the competition
[11].
Online technology collects large amounts of detailed data on visitor traffic and activities on
websites, which would cause a plethora of metrics [20], and on the other hand this variety of
measures can be overwhelming. Developing a website is a dynamic ongoing process which is
guided by knowledge of its visitors.
2. Profile of the website
In 1998 an Iranian visual artist website was launched (http://www.omidvar.net). This website
has many pages with images and few texts. The Google Analytics traffic overview showed that
all traffic sources sent a total of 19,703 visits from 1 June 2008 to 1 May 2010. The total page
views during this period were 100,144.
3. Methodology
Google Analytics allows users to export report data in Microsoft Excel format, which when
transformed can be analyzed with time series statistical programs. The software EViews is used
to compute time series regression. Initially a data set with 19,703 entries for 23 months drawn
from Google Analytics was employed to analyze the performance of page views or page views
per visits. Monthly data series was the most suitable series among daily and weekly because the
accuracy and credibility of the regression was higher than those of other series.
5. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
18
Figure 4 Independent Variables
6. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
19
Figure 5 Dependent Variables
Before the hypothesis testing, some statistical matters with regards to the use of Google
Analytics data in combination with time series methodology must be considered. For this reason
all independent variables were processed by Augmented Dickey-Fuller Test to see if they were
stationary or not. If the variables had a unit root they would be transformed to stationary by
Difference-Stationary Process.
Firstly, the total page views were broken down to its elements. For example page views of new
visitors and returning visitors are elements of total page views regarding to page views of
visitors type.
Secondly, Autoregressive Moving Average (ARMA) model were created with their independent
variables and their dependent variables.
Thirdly, a new ARMA model of total page views was created with the sum of its related
fundamental model. With the combined use of both dependent variables and independent
variables, more accurate results were achieved.
Finally, the new model credibility and reliability was checked which is consistent with these
seven steps [27].
7. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
20
1) Regression line must be fitted to data strongly (R^2>0.5).
2) Independent variables should be jointly significant to influence or explain the dependent
variable (i.e. F-test, Anova).
3) Most of the independent variables should be individually significant to explain
dependent variable (i.e. T-test).
4) The sign of the coefficients should follow economic theory or expectation or
experiences or intuition.
5) No serial or auto-correlation in the residual (Breusch-Godfrey serial correlation LM test
: BG test)
6) The variance of the residual (u) should be constant meaning that homoscedasticity
(Breusch-Pegan-Godfrey Test)
7) The residual (u) should be normally distributed (Jarque Bera statistics).
4. Hypothesis testing
The main questions are:
• Are page views well described by visitors’ sources? Which source has the highest
impact and why?
• Are page views well described by visitors’ connection speed? Which connection speed
has the highest impact and why?
• Are page views well described by visitors’ type? Which type has the highest impact and
why?
• This survey starts with a preparation stage which checks if our independent variables
are stationary series or not and transforms them to stationary.
Independent Variable Stationary Series Difference-Stationary Process
Search Visitors Yes
Direct Visitors No Yes
Referral Visitors Yes
New Visitors Yes
Returning Visitors Yes
Visitors with Unknown Speed Yes
Visitors with DSL Speed Yes
Visitors with Cable Speed Yes
Visitors with T1 Speed Yes
Visitors with Dial-Up Speed Yes
Visitors with OC3 Speed Yes
Table 1 Augmented Dickey Fuller (ADF) Test and Difference-Stationary Process
4.1 The impacts of visitors’ sources on page views:
As Table 1 illustrates, Direct Visitors are not stationary series therefore they are transformed to
stationary series by difference-stationary process and the new variable is called “Direct
Visitors*”.
In the Introduction it was mentioned that there are a variety of data in web analytics and the
metric values should be chosen wisely, therefore the independent variable (page view) is broken
down to the page views of direct visitors, referral visitors and search visitors as shown in
Table2.
8. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
21
Independent
Variable
Dependent
variable
R^2 Probe value Coefficient
Page views
of Search
Visitors
Search
Visitors
0.41 0 3.82
Page views
of direct
Visitors
Direct
Visitors*
0.3 0.01 4.44
Page views
of Referral
Visitors
Referral
Visitors
0.61 0 1.77
Table 2 ARMA Regression for Page Views Sources
Direct Visitors* was transformed with Difference-Stationary Process to stationary series,
therefore the amount of R^2 which is lower than 0.5 is not an issue to show the fitness of model
to data; but on the other hand, for search visitors which its R^2 is lower than 0.5 implies that the
regression is unlikely to be a good estimate.
Page views = page views of search engine visits + page views of direct visits + page views of
referral visits
Page views = 3.82*Search Visitors + 4.24*Direct Visitors + 1.77*Referral Visitors + c
Indepen
dent
Variabl
e
Dependen
t variables
Regressio
n Fitted to
Data
Strongly
Jointly
Significan
t
Coefficient
s Follow
Economic
Theories
No
Serial in
the
Residual
Homosc
edasticit
y of the
Residual
Varianc
e
Residual
Normali
ty
Distribu
ted
Total
page
views
Visitors
sources
Yes Yes Yes No Yes Yes
Table 3 Assumptions of good Regression model of total page views with visitors sources
Excepting the serial in the residual this regression is a good model and shows the impact of
Direct and Referral visitors correctly.
4.2 The impacts of visitors’ connection speed on page views:
The dependent variables of visitors connection speed are broken down and an ARMA module
for each is created as follows.
Independent
Variable
Dependent
variable
R^2 Probe value Coefficient
Page Views
of Visitors
with
Unknown
Speed
Visitors with
Unknown
Speed
0.69 0 4.81
Page Views
of Visitors
with DSL
Speed
Visitors with
DSL Speed
0.391 0 2.74
1
Unlikely to be a good estimate
9. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
22
Page Views
of Visitors
with Cable
Speed
Visitors with
Cable Speed
0.74 0 1.71
Page Views
of Visitors
with T1
Speed
Visitors with
T1 Speed
0.47 0 2.22
Page Views
of Visitors
with Dial-
Up Speed
Visitors with
Dial-Up
Speed
0.41 0 3.16
Page Views
of Visitors
with OC3
Speed
Visitors with
OC3 Speed
0.53 0 7.81
Table 4 ARMA Regression for Page Views Connection Speed
Visitors with DSL, T1 and Dial-Up speed are unlikely to be a good estimate because their R^2
is lower than 0.5.
Page views = page views of Unknown visitors + page views of DSL Visitors + page views of
Cable Visitors+ page views of Dial-Up Visitors+ page views of OC3 Visitors
Page Views=4.81*Unknown visitors +2.74*DSL Visitors +1.71* Cable
Visitors+2.22*T1+3.16*Dial-Up Visitors+7.81*OC3 Visitors +C
Independent
Variable
Dependent
variables
Regression
Fitted to
Data
Strongly
Jointly
Significant
Coefficients
Follow
Economic
Theories
No
Serial in
the
Residual
Homoscedasticity
of the Residual
Variance
Residual
Normality
Distributed
Total page
views
Visitors
Connection
Speed
yes yes yes yes yes yes
Table 5 Assumptions of good Regression model of total page views with visitors Connection
Speed
4.3 The impact of visitors’ type on page views:
Independent
Variable
Dependent
variable
R^2 Probe value Coefficient
Page Views
of New
Visitors
New
Visitors
0.68 0 2.40
Page Views
of
Returning
Visitors
Returning
Visitors
0.70 0 5.22
Table 6 ARMA Regression for Page Views Type
Page Views= Page Views of New visitors+ Page Views of Returning Visitors
Page Views= 2.40*New visitors+ 5.22*Returning Visitors
10. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
23
Independe
nt Variable
Dependent
variables
Regression
Fitted to
Data
Strongly
Jointly
Significant
Coefficients
Follow
Economic
Theories
No Serial
in the
Residual
Homosce
dasticity
of the
Residual
Variance
Residual
Normality
Distributed
Total page
views
Visitors
Connection
Speed
yes yes yes yes yes yes
Table 7 Assumptions of good Regression model of total page views with visitors Type
5. Results and Analysis
The results of this survey are categorized in two categories:
1. Results from Regression
2. The methodology that is used
5.1 Results from Regression:
Most of the direct visitors have visited the site before and memorized the address or entered the
site from traditional channels so namely they could be fan of the website, the website owners
and the website developers or visitors from traditional channels (conference, meeting, and
event). Those visitors who access the website from traditional channels would probably increase
during the time of event or meeting and those visitors who are site owners or developers will
probably increase visits during the update or develop time. Consequently these visitors were not
stationary and had unit root.
Analyzing the impact of visitors sources on pages views
By each direct visitor, the number of page views was increased to 4.44 and for referral visitors it
was increased to 1.77; therefore the direct visitors had a higher impact in comparison to referral
visitors. Additionally, exchanging links with more related websites do improve referral visitors’
effect on page views. The Page views of search visitors were not fitted to the module and varied
very much. In order to understand search visitors behavior, their quarries meaning should be
understood which is available through semantic web technologies. Investigation of keywords’
priority and bases on discarded certain percentage of web pages hence making the search
more compact and efficient [25].
Analyzing the impact of visitors connection speed on pages views
By literature review it was thought that the higher speed connection would cause a greater
impact on page views, but things are not so simple. This is due to access by Dial-Up visitors
which used a lower connection speed and had more impact on page views than cable and T1
visitors. It is believed that the content type on the region of visitors plays an important role on
that. Since this site belongs to an Iranian artist, it must have a higher impact on visitors from
Iran.
11. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
24
Figure 6 Percentage of Territory of Unknown visitors
Figure 6 reveal that Unknown visitors are mainly from Iran therefore it proves their high impact
on page views.
Figure 7 percentage of Territory of DSL Visitors
17% of DSL visitors which on average had a speed of 1.5 Mbps were from Iran. That is the
reason why it had more impact on page views than Cable visitors with average speed of 4 Mbps.
Figure 8 Percentage of territory of Cable visitors
12. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
25
Cable visitors had the lowest impact although their average speed was around 4 Mbps. This can
be explained by the low percentage of Iranian visitors.
Figure 9 Percentage of Territory of T1 visitors
20.21% of T1 visitors are from Iran and the connection speed roughly would be 1.5 MBPS. It is
expected that these visitors had the same impact as DSL visitors; and, indeed their impact was
close to each other (2.22 for T1 visitors and 2.74 for DSL visitors).
Figure 10 Percentage of Territory of Dial-Up visitors
Dial-Up visitors are mainly from Iran so it does prove their impact (3.16) on page views. The
Oc3 visitors had the highest speed which on average is 155 MBPS; therefore their impact on
page views was as high as 7.81.
In conclusion, what can be manifest from close analysis of these data is that it is easier to
predict user behavior whose territory and connection speed does not varied much.
Analyzing the impact of visitors Type on pages views
Returning visitors had more impact on page views than new visitors did and it was quite high.
Returning visitors must have similarities with direct visitors because all of the direct visitors
excepting visitors from traditional channels had visited the site before. So their impact must be
close to each other (5.22 for returning visitors and 4.24 for direct visitors).
13. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
26
5.2 The New Methodology
Many websites have measured their success with their competitors’ behavior and many others
have measured that with their own website success. This survey had introduced a methodology
to measure the success of the website with its time series data. It also focused on one of the most
primary and important variables which are page views and showed how to use the most suitable
data for that. This method can be used on all websites and time series variables.
Recommendation
It is suggested to use this methodology on some other artists’ sites, and other variables such as a
combination of these variables. Furthermore since Google Analytics is integrated with Adwords
and Adsense and their time series data’s are available on Google Analytics, further studies on
time series data of Adsense Revenue or Adwords campaigns might have interesting results;
because they bring visitors and revenue to the site.
This methodology is not recommended for search visitors’ behavior and needs further
investigation. In order to understand search visitors behavior their quarries meaning should be
understood which is available through semantic web technologies. Investigation of keywords’
priority and classification would make the search more compact and efficient [25].
According to user referral link, territory and the meaning of the content, the ontology can be
simulated by Protégé which is a free, open source platform for constructing, visualizing and
manipulating domain models and knowledge based applications with ontologies [35].
Discussion
Many website owners or developers have used web traffic elements for their website
performance, but the important thing is the knowledge gained about visitors and their behavior
to keep them happy and satisfied with the website. Most of the websites have used web
analytics to collect data about visitors with no systematic way to convert these data into tangible
knowledge. The amount of these data which is available through web analytic is immense, and
the developer may get lost in it. So a systematic way is needed to analyze these data.
Firstly, this survey used time series regression with dependent variables and independent
variables. And all these variables were chosen wisely to have the most accurate result.
Secondly, the total page views module was created with sum of detailed regression module. For
example for new and returning visitors, the total page views were created as follow:
Page views of New visitors= New visitors*2.40 + C
Page views of Returning visitors= Returning visitors*5.22 + C
Total page views= Page views of new visitors + Page views of Returning visitors + C
Finally, seven assumptions as listed by Hossain were checked as the final module to see how
good the ARMA module was [28].
In sum, this paper showed how to use a regression module with the most suitable dependent and
independent variables and its functionality.
APPENDICES
Augmented Dickey-Fuller Unit Root Tests:
The variables under the study are the following: Direct visits (DV), reference site visits (RSV),
search engine visits (SV), new visits (NV), returning visits (RV), visitors with unknown speed
(UNKNOWNV), visitors with DSL Speed (DSLV), visitors with Cable Speed
14. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
27
(CABLEV),visitors with T1 Speed (T1V), visitors with Dial-Up Speed (DIAL-UPV) and
visitors with OC3 Speed (OC3V).
DV ADF Test Statistic:-2.85 10 percent Critical Value:-2.65
RSV ADF Test Statistic:-2.15 10 percent Critical Value:-2.64
SV ADF Test Statistic:-2.85 10 percent Critical Value:-2.65
SV* ADF Test Statistic:-4.48 10 percent Critical Value:-2.65
NV ADF Test Statistic:-3.95 10 percent Critical Value:-2.64
RV ADF Test Statistic:-2.74 10 percent Critical Value:-2.64
UNKNOWNV ADF Test Statistic:-2.73 10 percent Critical Value:-2.64
DSLV ADF Test Statistic:-3.62 10 percent Critical Value:-2.64
CABLEV ADF Test Statistic:-4.51 10 percent Critical Value:-2.64
T1V ADF Test Statistic:-3.65 10 percent Critical Value:-2.64
DIAL-UPV ADF Test Statistic:-3.49 10 percent Critical Value:-2.64
OC3V ADF Test Statistic:-2.2 10 percent Critical Value:-1.60
ARMA Regression:
Dependent Variable: PAGEVIEWS
Method: Least Squares
Date: 10/06/10 Time: 12:28
Sample (adjusted): 2008M07 2010M04
Included observations: 22 after adjustments
Page views=3.82*Search Visitors+4.24*Direct
Visitors+1.77*Referral Visitors+ c
Coefficient Std. Error t-Statistic Prob.
C(1) 2609.016 256.2849 10.18014 0.0000
R-squared 0.581917 Mean dependent var 4252.682
Adjusted R-squared 0.581917 S.D. dependent var 1859.102
S.E. of regression 1202.083 Akaike info criterion 17.06589
Sum squared resid 30345061 Schwarz criterion 17.11548
Log likelihood -186.7248 Hannan-Quinn criter. 17.07757
Durbin-Watson stat 0.947281
Table 8 ARMA Regression for Sources
15. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
28
Dependent Variable: PAGEVIEWS
Method: Least Squares
Date: 10/07/10 Time: 20:02
Sample (adjusted): 2008M06 2010M04
Included observations: 23 after adjustments
Page Views=4.81*Unknown visitors +2.74*DSL
Visitors +1.71* Cable
Visitors+2.22*T1+3.16*Dial-Up
Visitors+7.81*OC3 Visitors +C
Coefficient Std. Error t-Statistic Prob.
C(1) 1332.591 198.0689 6.727918 0.0000
R-squared 0.742444 Mean dependent var 4346.913
Adjusted R-squared 0.742444 S.D. dependent var 1871.734
S.E. of regression 949.9051 Akaike info criterion 16.59311
Sum squared resid 19851032 Schwarz criterion 16.64248
Log likelihood -189.8207 Hannan-Quinn criter. 16.60552
Durbin-Watson stat 1.560109
Table 9 ARMA Regression for connection speed
Dependent Variable: PAGEVIEWS
Method: Least Squares
Date: 06/28/10 Time: 21:54
Sample (adjusted): 2008M06 2010M04
Included observations: 23 after adjustments
PAGEVIEWS=2.4*A__+5.22*B__+C(1)
Coefficient Std. Error t-Statistic Prob.
C(1) 1843.713 212.7060 8.667892 0.0000
R-squared 0.702971 Mean dependent var 4346.913
Adjusted R-squared 0.702971 S.D. dependent var 1871.734
S.E. of regression 1020.102 Akaike info criterion 16.73570
Sum squared resid 22893393 Schwarz criterion 16.78507
Log likelihood -191.4605 Hannan-Quinn criter. 16.74811
Durbin-Watson stat 1.549155
Table 10 ARMA Regression for Types
Breusch-Godfrey Serial Correlation LM Test (BG Test)
Breusch-Godfrey Serial Correlation LM Test:
F-statistic 3.558785 Prob. F(2,19) 0.0487
Obs*R-squared 5.995448 Prob. Chi-Square(2) 0.0499
Table 11LM Test on ARMA model of Sources
16. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
29
Breusch-Godfrey Serial Correlation LM Test:
F-statistic 1.115946 Prob. F(2,20) 0.3472
Obs*R-squared 2.309003 Prob. Chi-Square(2) 0.3152
Table 12 LM Test on ARMA model of Connection speed
Breusch-Godfrey Serial Correlation LM Test:
F-statistic 1.293293 Prob. F(2,20) 0.2963
Obs*R-squared 2.633929 Prob. Chi-Square(2) 0.2679
Table 13LM Test on ARMA model of Type
Breusch-Pegan-Godfrey Test:
Heteroskedasticity Test: Breusch-Pagan-Godfrey
F-statistic 0.065414 Prob. F(3,18) 0.9775
Obs*R-squared 0.237265 Prob. Chi-Square(3) 0.9714
Scaled explained SS 0.288654 Prob. Chi-Square(3) 0.9621
Table 14Breusch-Pagan-Godfrey test on model of sources
Heteroskedasticity Test: Breusch-Pagan-Godfrey
F-statistic 2.066530 Prob. F(6,16) 0.1153
Obs*R-squared 10.04188 Prob. Chi-Square(6) 0.1229
Scaled explained SS 9.298063 Prob. Chi-Square(6) 0.1575
Table 15Breusch-Pagan-Godfrey test on model of Connection Speed
Heteroskedasticity Test: Breusch-Pagan-Godfrey
F-statistic 1.060743 Prob. F(2,20) 0.3649
Obs*R-squared 2.205738 Prob. Chi-Square(2) 0.3319
Scaled explained SS 1.809690 Prob. Chi-Square(2) 0.4046
Table 16Breusch-Pagan-Godfrey test on model of Types
Normality Test for the residuals
Jarque-Bera Test for visitors’ Source: 3.92 Prob: 0.14
Jarque-Bera Test for visitors’ Connection Speed: 3.29 Prob: 0.19
Jarque-Bera Test for visitors’ Type: 2.48 Prob: 0.29
17. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
30
References
[1] Analytics Usage Statistics. (2010, July). Retrieved from BuiltWith Technology Usage Statistics:
http://trends.builtwith.com/analytics
[2] Cookies & Google Analytics. (2010). Retrieved 2010, from google:
http://code.google.com/apis/analytics/docs/concepts/gaConceptsCookies.html
[3] Google Analytics Usage Statistics. (2010, Jun). Retrieved from Web and Internet Technology
Usage Statistics: http://trends.builtwith.com/analytics/Google-Analytics
[4] Association, W. A. (n.d.). “Big Three Definitions” Ver. 1.0. 2300 M Street, Suite 800,
Washington DC 20037, United States.
[5] Association, W. A. (2006). Web Analytics “Big Three” Definitions. Washington DC 20037.
[6] Assumptions of Linear Regression. (n.d.). Retrieved July 2010, from homepages.ius.edu:
http://homepages.ius.edu/WCLANG/m305/notes11.htm
[7] B.king, A. (2008). Website Optimization. Orielly.
[8] Barua, A., Konana, P., B.Whinston, A., & yin, F. (2001.). Driving e-business excellence. Sloan
Management, Rev. 34(1) 36–44.
[9] Batchelor, P. R. (n.d.). EVIEWS tutorial:Cointegration and error correction. City University
Business School, London.
[10] Beatriz Plaza Faculty of Economics, U. o. (2009). Monitoring web traffic source effectiveness
with Google Analytics An experiment with time series. Emerald, 9.
[11] Bking, A. (2008). Website optimization. Oreilly.
[12] Bodily, S., & Venkataraman, S. (2004). Not walls, windows: capturing value in the digital age.
Journal of Business Strategy, Vol. 25 No. 3, pp.15-25.
[13] capital, I. u. (n.d.). ITU World Telecommunication/ICT indicators database.
[14] Chakraborty, G. L. (2002). An empirical investigation of antecedents of B2B Web sites’
effectiveness. Journal of Interactive Marketing,Vol. 16 No. 4, 51-72.
[15] Chakraborty, G. L. (2002). An empirical investigation of antecedents of B2B Websites'
effectiveness. Journal of Interactive Marketing, 16: 51–72. doi: 10.1002/dir.10044.
[16] Consotium, I. S. (2010). The ISC Domain Survey | Internet Systems Consortium. Retrieved from
Internet Systems Consortium | Internet Systems Consortium: http://www.isc.org/solutions/survey
[17] Cotter, S. (1993). TAKING THE MEASURE OF E-MARKETING SUCCESS. Journal of
Business Strategy, Vol. 23 Iss: 2, pp.30 - 37.
[18] Eisenberg, B. ( November 26, 2007). Future Now’s 2007 Retail Customer Experience Study.
[19] Eugenia Y. Huang, D. o. (2005). Is revamping your web site worthwhile? Emerald, 15.
[20] FMI Group. (2001). Web site Visitors Analysis-Statistics or Intelligence? Basinstoke: FMI
Group.
[21] Fourie, I., & Bothma, T. (2007). Information seeking: an overview of web tracking and the
criteria for tracking software. Aslib Proceedings, ISSN: 0001-253X, 24.
[22] Garson, G. D. (20010,2009,2008). Logistic Regression. Retrieved July 2010, from
http://faculty.chass.ncsu.edu: http://faculty.chass.ncsu.edu/garson/PA765/logistic.htm
[23] Global Number of Internet Users,total and per 100 inhabitants 2000-2009. (n.d.). Retrieved from
ITU: Committed to connecting the world: http://www.itu.int/ITU-
D/ict/statistics/material/graphs/Internet_users_00-09.jpg
[24] Google, A. c. (2009). Appraising Your Investment In Enterprise Web Analytics. Forrester
consulting, (p. 20).
18. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
31
[25] Gupta, Siddharth; Thakur, Narina;. (2010). Semantic Query Optimisation with Ontology.
International journal of Web & Semantic Technology (IJWesT).
[26] Hoffman, D. L., & Novak, T. P. (October 1996). Marketing in Hypermedia Computer-Mediated
Environments. journal of marketing, 50-68.
[27] Hossain, S. (n.d.). An Investigation into Regression Model using EVIEWS. Lecturer for
Economics.
[28] Hossain, S. (n.d.). An Investigation into Regression Model using EVIEWS. Retrieved from
http://www.sayedhossain.com/: www.sayedhossain.com/files/Lec1.Regression.ppt
[29] http://www.hkbu.edu.hk/~billhung/econ3600/application/app01/app01.html. (n.d.). Dickey-
Fuller Unit Root Test. Retrieved from Hong Kong baptist university:
http://www.hkbu.edu.hk/~billhung/econ3600/application/app01/app01.html
[30] Inge Geyskens, K. G. (2002). The Market Valuation of Internet Channel Additions. The Journal
of Marketing, Vol. 66, No. 2 (Apr., 2002), pp. 102-119 .
[31] Jacobs, I. (2002, May). DRAFT: Web Achitecture Document. Retrieved from World Wild Web
Consortium: http://www.w3.org/2001/tag/2002/0508-intro
[32] Kohavi, R. e. (2007). Practical Guide to Controlled Experiments on the Web: Listen to Your
Customers.
[33] Köhler, N. V. (1999, Sep). Measuring Internet User Traffic Behavior Dependent on Access
Speed. Institute of Computer Science, Germany.
[34] Lederer , A., Mirchandani, D., & Sims, K. (2001). The Search for Strategic Advantage from the
World Wide Web. International Journal of Electronic Commerce, Volume 5 , Issue 4 Pages:
117-133 .
[35] Li BAI, Min LIU, “A Fuzzy –set based Semantic Similarity Matching Algorithm for Web-
Services” CIMS Research Centre, IEEE 2008
[36] Ludlow, E. (n.d.). Multiple Regression: Fitting Models for Multiple Independent Variables.
[37] MArk F.Taylor, D. E. (2001). Knowledge for web site development. Emerald, 11.
[38] Okazaki, S., & Rivas, J. A. (2002). A content analysis of multinationals’ Web communication
strategies: cross-cultural research framework and pre-testing. Internet Research, Vol. 12 No. 5,
pp.380-90.
[39] Sarah Quinton and Mohammed Ali Khan, B. S. (2009). Generating web site traffic: a new model
for SMEs. emerald, 15.
[40] Schonfeld, E. (2010, March 8). Forrester Forecast: Online Retail Sales Will Grow To $250
Billion By 2014. Retrieved from TechCrunch: http://techcrunch.com/2010/03/08/forrester-
forecast-online-retail-sales-will-grow-to-250-billion-by-2014/
[41] Search Engine Marketing Professional Organization, S. (2008, February). Search engine
marketing 2007.
[42] SEMPO. (2006). Search Engine Marketing Professional Organization survey of SEM agencies
and advertisers.
[43] Thompson, K. K. (n.d.). Secure 8000/month wit Adsense arbitrage.
[44] Welling , R., & WhiteMacquarie , L. (2006). Web site performance measurement: promise and
reality. Managing Service Quality, SSN: 0960-4529.
19. International Journal of Web & Semantic Technology (IJWesT) Vol.2, No.1, January 2011
32
Authors
Mohammad Amin Omidvar received his master information technology
management degree from Islamic Azad University (E-campus), Tehran,
Iran, in 2010. He has had many researches in E-commerce field and as an
innovator he has deployed information technology in various fields
especially web, multimedia and digital painting which has been presented
in many Iranian universities’ speeches and conferences.
Vahid Reza Mirabi has PH.D in marketing management and is working as
a vice president of management school and was the president of Islamic
Azad university Tehran central branch. He has more than 16 years of
experience in academic and industry in Iran. He received his PH.D from
GHARWAL Srinegar from Poona University in India. He is advisor of
many companies in the field of marketing. He has published 5 books and
several articles. His major area of interest is marketing research, strategic
management and human resource management.
Narjes Shokri is working as a head department of public administration
(management) and has more than 10 years of experience in academics and
more than 15 years of experience in teaching. She received BA and MA
in public administration at Tehran University and PH.D at Islamic Azad
University Tehran. Her major area of interest is method of research, MIS
and organizational behaviors. She has authored several researches in
national journal.