First Author\[1\]\*, Second Author2 (Double Blind Review: Please do not type or edit anything here until final-camera ready submissions) 1First affiliation, City and Country (Please do not type or edit anything here, our editors will do the work for you) 2Second affiliation, City and Country (Please do not type or edit anything here, our editors will do the work for you)
ARTICLE INFO ABSTRACT

Article history:

Received XX Month 2024

Revised XX Month 2024

Accepted XX Month 2024

Online first

Published 1 September 2024

 Nowadays it is critical to understand the behaviour of consumers due to various demographic background making consumer to behave differently. Running a business is now tough due to various factors such as competition in business and online shopping platform that offers cheaper price. The help of big data technologies can help businesses to gain insights on consumer behaviour of online shopping. This study suggested the development of data visualization of consumer financial online spending (CFOS) using Power BI tool which can help businesses to analyse consumer behaviour as well as make data-driven decision making. The agile software development method was applied where it consists of requirement analysis, design, implementation, testing and evaluation phase. Selected consumer behaviour dataset is derived from Kaggle. Then, Jupyter Notebook is used to perform ETL process and data cleaning on the dataset to eliminate bias leading to invalid conclusions and visualizations. Power BI tools is used to visualize and develop the dashboard with various charts and graphs, then it is embedded into CFOS website. The functional testing and usability testing are carried out to evaluate the developed CFOS website. There were 31 respondents who participated in testing the usability of dashboard. This study outcome indicates positive feedback on CFOS website, as well as for the integration of the dashboard. This calls for future works to improve the current CFOS website and dashboard to be more flexibility, ease to use and more user friendly.

Keywords:

Big data

Consumer behaviour

Spending habits

Data Visualization

Dashboard

Business Intelligence

DOI:

10.24191/jcrinn.v9i2
# INTRODUCTION Consumers, particularly employees, have a significant role in improving a country's economic status. They are active in financial operations and decision-making daily until they retire. They play an important role in the success of a business and sales. Consumers from all demographics behave differently when it comes to online spending. Understanding these behaviours, as well as measuring the influence of various demographic parameters such as age, gender, and education on online spending, is critical. Investors, businesses, and policymakers closely follow published statistics and reports on consumer spending in order to help forecast and plan investment and policy decisions. ## Problem Statement Consumer spending is, naturally, crucial to businesses. The more money consumers spend at a given business, the better that business tends to perform. For this reason, it is unsurprising that most investors and businesses pay a great amount of attention to consumer spending figures and patterns. Investors and businesses closely follow consumer spending statistics when making forecasts. However, businesses do not have large enough data and lack of tools, therefore, there is no dashboard in analysing consumer spending behaviour accurately and precisely. This led to decreasing in sales performance and poor consumer satisfaction level (Das, 2022). This situation has been known to cause poor revenue due to unwise investments on marketing (ThinkSecure Network, 2021). Businesses may develop and implement marketing strategies that are in line with these groups of consumers by understanding the target consumers' access intensity (Rosário & Raimundo, 2021). If the businesses continue with the current trend in marketing strategy and no usage of dashboard, they will be in danger of compromising their overall sales performance and business revenue (Arora, 2023). Without a dashboard, to help businesses achieve its target sales without investing lots of time and cost, salesmen will continue promoting their business goods and services inefficiently and produce minimal profit in a long amount of time which can slow the business growth (Xeo Blog, 2022). Siti et al. (2019) discovered that to appeal to target consumers, make it relatable, and pique their interest in making a purchase, effective marketing material should be more visual and add a human touch. To increase sales performance in the future, a study on consumers spending habits or behaviour must be conducted to determine the patterns of consumer when spending their money on a goods. ## Objectives This study aims to investigate the consumers spending behaviour that can reveal insights into different consumers segments based on their demographic characteristics to better target future marketing strategy. This analysis may assist the businesses to prevent pointless money spending on marketing strategy and improve sales performance. The objectives of this study are: 1) To identify data analysis requirements for consumer financial online spending behaviour on goods and services. 2) To design and develop a data visualisation dashboard based on processed data. 3) To evaluate both functionality and usability of the developed dashboard. # LITERATURE REVIEW ## The importance of consumer spending Consumer spending helps in forecasting the economy performance (Indeed, 2022). Consumer spending is the main engine of the country’s economy, and business, government, and consumer financial activity all contribute to national economies (Chron, 2020). The national economy may suffer because of consumers cutting back on their consumption when they increase their savings, investments, or debt repayments. Data on consumer spending from prior years may be used by economists to predict how the economy will function in the future. Accurate forecasts that consider consumers' upcoming consumption needs enable owners and management of businesses to maximise and prolong revenues throughout time. Consumer spending helps businesses adapt to the new changes (Indeed, 2022). Consumer spending may be used by businesses to adapt to shifting consumer preferences. They can continue to attract clients and increase revenue by being able to adapt. Consumer spending helps in increasing market competition between businesses with similar goods and services (Indeed, 2022). Data on consumer spending may boost rivalry between comparable businesses, especially if there is a sharp increase in demand for specific goods and services. Data on consumer spending assists businesses in identifying the goods that are most valuable in the market (Chron, 2020). ## Factors influencing consumer behaviour The ultimate purchase choice is significantly influenced by the consumer's purchasing intention. Additionally, information has a significant role in influencing consumers' intent to purchase as well as their final choice. Consumers are often risk-averse, thus they will gather a lot of pertinent information prior to making a purchase to convert the uncertainty of purchasing a certain goods into certainty (Tao et al., 2022). The consumer behaviour when online shopping and offline shopping posed a difference. Consumers prefer to choose online shopping to avoid the hassle of driving themselves to the physical shop and it is convenient and quick to make a purchase anytime and anywhere. Besides, online shopping has the characteristics of diversity and portability (Dong, 2022). Social factors, cultural factors, demographic factors, and situational factors all have an impact on changes in consumer purchase behaviour (Cici & Bilginer, 2021). ## Visualization Dashboard Siti et al. (2019) discovered that to appeal to target consumers, make the product relatable, and pique their interest in making a purchase, effective marketing material should be more visual and add a human touch. Big data analysis is the process of analysing massive amounts of data, which may be summed up as five Vs, featuring a large volume, velocity, variety, value, and veracity (Dong, 2022). Big data, which has been heavily utilised in the field of marketing, is crucial for the analysis of consumer behaviour, for predicting potential future changes in consumer needs and wants, and for creating marketing strategies that are appropriate for these desires and needs (Degermen & Mohammadabbasi, 2023). Data visualisation involves placing data or information into a pictorial or graphical environment, such as a chart, a map, or other visual forms to make information easier for the human brain to perceive and interpret (Brush & Burns, 2022). Businesses may visualise data using dashboards by fusing a variety of graphs, charts, and other data visualisation widgets. The dashboard can display data regarding consumer online spending as a chart, graph, or diagram. Inappropriately choosing data visualisation charts and graphs are contributing factor in dashboard design errors. Understanding the function of a chart, which is basically to make it simple for users to spot patterns and compare numbers to one another, is necessary when choosing charts (Orlovskyi & Kopp, 2020). The created dashboard may mislead businesses and divert focus and attention to insignificant or unimportant information if improper visualisation charts are selected that do not match the nature of the data contained in datasets prepared for visualisation (Orlovskyi & Kopp, 2020). # METHODOLOGY ## Agile methodology Agile development model is an incremental developmental model. Software is developed in many iterative cycles. This study aims to develop an interactive dashboard with the visualization of consumer financial online spending data for the businesses. The agile development methodology is chosen in this research. It adopts the dashboard development process throughout the development phase. The dashboard development process consists of five main phases, which are the requirements analysis phase, design phase, implementation phase, testing phase and evaluation. In the requirement analysis phase, information or data related to research problems were collected. A literature review approach was used to obtain such information. Gathering sources of literature as references to support the topic and serves as a more convincing theoretical foundation. The reading sources that can be used as references are in the form of soft copy reading sources obtained from the Internet. Design phase means how to design a system that will be built based on the results of previous analysis. At this phase, required concepts, user interface and functions are designed. Data visualization of consumer financial online spending behavioral analysis which will process raw data of consumer online spending to produce a dashboard about the spending behaviour. The designed dashboard in previous phase is developed in implementation phase. Testing and evaluation phase refers to the testing the performance of a dashboard that has been developed by using both functionality and usability testing method to determine whether the dashboard designed is feasible to use. The outcome of the conducted testing then is evaluated for improvement purposes. Figure 1 shows the Agile methodology used in this study. ![](6682a7232b3ee_media/media/image1.emf) Figure 1 Agile methodology ## Extract, Transform and Load (ETL) As part of the implementation phase, there are three sections in data cleaning process, which are extract, transform, and load, which aims at the systematic processing of source data to make it available in a format more convenient for the intended use. Data cleaning is the process of fixing or removing incorrect, corrupted, incorrectly formatted, duplicate, or incomplete data within a dataset. In extract, raw data is transferred or exported from source locations to a staging area during data extraction. A range of data sources, both structured and unstructured, are used to extract data. In transform, the staging area is where the raw data is processed. For its intended analytical use case, the data is changed and consolidated in this place. In load, the converted data is sent from the staging area into the target data warehouse in this final stage. This often entails initial loading of all data, recurring loading of incremental changes to the data, and, less frequently, full refreshes to completely remove and replace all data in the warehouse. During data cleaning in this study, a few modifications have been made. 1) Set the name of the columns to all lowercase. 4) Drop Timestamp column as they are not relevant in this analysis for simplicity. 5) Drop values that do not satisfy logical condition since it is impossible for 3-years-old to answer the survey based on the dataset. 6) Fill two missing values in product search method column with mode, which is categories. 7) Create a new column of polarity shopping satisfaction then assign either positive or neutral or negative to the rating. 8) Split age into different age groups then create a new column named age category with the grouped age range. 9) Split purchase categories column with a list type value to multiple rows. The first step required in the data cleaning process is to import compulsory packages. Figure 2 shows an example of the query used to import packages for data cleaning process using Jupyter Notebook. ![A screenshot of a computer Description automatically generated](6682a7232b3ee_media/media/image2.png) Figure 2 Query used to import packages for data cleaning process There are plenty of other queries within the transform process, including the query for identifying the source of the null values and missing data. It is crucial in ETL process because it can reduce the statistical power of a study and can produce biased estimates, leading to invalid conclusions. Figure 3 shows an example of how a query is used to show the number of missing values in the data frame by columns. ![A screenshot of a computer Description automatically generated](6682a7232b3ee_media/media/image3.png) Figure 3 Query used to show the number of missing values in the data frame by columns After the process of cleaning the data is completed, the processed data frame is saved into a csv file before being loaded and used in dashboard development using Microsoft Power BI software. Figure 4 shows the overview page of the CFOS dashboard that visualizes demographic of a user. Then after dashboard building process is completed, it is integrated as part of the CFOS website. ![](6682a7232b3ee_media/media/image4.png) Figure 4 Overview page of the CFOS dashboard that visualizes demographic of a user # RESULT AND DISCUSSION ## Functional testing The method used for functional testing is applied on the administration side of the CFOS website by making sure that each stated functional requirement is tested. A test case is used to verify the status of each test case if the expected output produced the same result as the actual output. If the expected output and actual output are the same, the status is success; otherwise, the status is failed. Table 1 shows the test case of functional testing. **Table 1** Test case of functional testing | **No** | **Test Case ID** | **Description** | **Expected Output** | **Actual Output** | **Status** | | ------ | ---------------- | ------------------------------------------------------------------- | -------------------------------------------------------------------- | -------------------------------------------------------------------- | ---------- | | 1 | TC01 | Admin updates profile information | View profile page displayed with new updated profile information | View profile page displayed with new updated profile information | Success | | 2 | TC02 | Admin adds new news on online shopping scams | View news page displayed with new added news online shopping scams | View news page displayed with new added news online shopping scams | Success | | 3 | TC03 | Admin updates information on existing news on online shopping scams | View news page displayed with new updated news online shopping scams | View news page displayed with new updated news online shopping scams | Success | | 4 | TC04 | Admin deletes existing news on online shopping scams | View news page displayed with updated news minus the deleted one | View news page displayed with updated news minus the deleted one | Success | ## Usability Testing After experiencing the look and feel of the CFOS website, respondents are required to answer the Post-Study System Usability Questionnaire (PSSUQ) with standard 16 questions used to evaluate user’s perception and satisfaction of a website, system or a product. Table 2 shows the result of usability testing questions. The Likert scale is used to measure respondent’s opinions on each statement, ranging from 1 to 7, with 1 being strongly agree and 7 being strongly disagree. The mean score for each statement is calculated. The average mean score is 2.61 to 3.74. **Table 2** Usability testing result (PSSUQ)
No Post Study System Usability Questionnaire Mean
System Usefulness
1. Overall I am satisfied with how easy it is to use this system. 3.09
2. It was simple to use this system. 3.03
3. I was able to complete the tasks and scenarios quickly using this system. 3.41
4. I felt comfortable using this system. 3.29
5. It was easy to learn to use this system. 3.12
6. I believe I could become productive quickly by using this system. 3.58
Information Quality
7. The system gave error messages that clearly told me how to fix problems. 3.74
8. Whenever I made a mistake using the system, I could recover easily and quickly. 3.51
9.

The information (such as online help, on-screen messages, and other

documentation) provided by the system was clear.

 3.48
10. It was easy for me to find the information I needed. 3.35
11. The information was effective in helping me complete the tasks and scenarios. 3.45
12. The arrangement of information on the system screens was clear. 3.45
Interface Quality
13. The interface of this system was pleasant. 3.41
14. I liked using the interface of this system. 3.41
15. This system has all the functions and capabilities I expect it to have. 3.32
16. Overall, I am satisfied with this system. 2.61
Based on the conducted usability testing, which was answered by 31 respondents, the mean score for each criterion of usability is calculated and compared with the recommended score. The recommended score is based on the means determined by Sauro and Lewis (2012) which involving 21 studies and 210 participants. The lower the mean score, the better the performance and satisfaction. Therefore, the CFOS website and dashboard usability score is on average. Table 3 shows the CFOS usability testing score comparison between the recommended score and actual score. The overall score is the average mean score of questions 1 to 16. The system usefulness is determined by the average mean scores of questions 1 to 6. The information quality is the average mean score of questions 7 to 12. The interface quality is the average mean score of questions 13 to 15. **Table 3** Usability testing score | **No** | **Criteria** | **Recommended score** | **Actual score** | | ------ | ------------------- | --------------------- | ---------------- | | 1\. | Overall | 2.82 | 3.32 | | 2\. | System usefulness | 2.80 | 3.25 | | 3\. | Information quality | 3.02 | 3.49 | | 4\. | Interface quality | 2.49 | 3.38 | Figure 5 shows the bar chart of the mean score of the usability testing. It shows that the average mean of the usability testing score suggests that the respondents agree that the CFOS website in terms of overall, system usefulness, information quality, and interface quality is usable. ![](6682a7232b3ee_media/media/image6.png) Figure 5 Usability testing score bar chart # CONCLUSION This study was undertaken to design and develop a visualization dashboard for consumer financial online spending behavioural analysis and evaluate the application by utilizing the usability testing and functional testing. The development of CFOS website and dashboard using Power BI has successfully completed to achieve its objectives. The CFOS website and dashboard will help the user which is businesses in increasing or boosting their sales more easily and effectively based on consumer behaviour. Although this CFOS website and dashboard had successfully developed and offers the user a great deal of value, improving the dashboard interface and CFOS website and dashboard will be crucial for future work. Among few limitations of CFOS website, based on respondents’ feedback includes being less flexible and not user friendly enough. In addition, the current CFOS website and dashboard are only perfectly adjustable to website view only. Finally, most respondents are keen of the idea to add support features that can assist the user if they have any inquiries or suggestions to developer. The support feature also acts as a customer service support for user to reach out to developer easily. # Acknowledgements/Funding (DOUBLE-BLIND reviewing. Leave the section as is. Only include Acknowledgements text in the final submission paper) This section is compulsory. The following is an example of an acknowledgement statement: The authors would like to acknowledge the support of Universiti Teknologi Mara (UiTM), Cawangan Negeri Sembilan, Kampus Kuala Pilah and Faculty of Applied Sciences, Universiti Teknologi MARA, Shah Alam, Selangor, Malaysia for providing the facilities and financial support on this research. # Conflict of interest statement (DOUBLE-BLIND reviewing. Leave the section as is. Only include CONFLICT OF INTERESTS text in the final submission paper) # This section is compulsory. The following is an example of a conflict-of-interest statement: The authors agree that this research was conducted in the absence of any self-benefits, commercial or financial conflicts and declare the absence of conflicting interests with the funders. # Authors’ contributions Each author contribution must be stated clearly reflecting each contribution to the body of the work and manuscript. Authors can refer to [CRediT](http://credit.niso.org/) (Contribution Roles Taxonomy) for the detailed information about individual contributions to the work. For example ***(Double Blind Review: Leave this section blank until final camera-ready submission)*:** **Anis Muneerah Shaiful Bahari**: Conceptualisation, methodology, formal analysis, investigation and writing-original draft; **Nurhaswani Alias**: Conceptualisation, methodology, and formal analysis; **Zainovia Lockman**: Conceptualisation, formal analysis, and validation; **Haslina Misran**: Conceptualisation, supervision, writing- review and editing, and validation. # References Arora, R. (2023). What is a business dashboard and why is it important? LeadSquared. [https://www.leadsquared.com/learn/sales/what-is-a-business- dashboard/](https://www.leadsquared.com/learn/sales/what-is-a-business-%20dashboard/) Brush, K., & Burns, E. (2022). What is data visualization and why is it important?. TechTarget. https://searchbusinessanalytics.techtarget.com/definition/datavisualization Chron. (2020). The importance of consumer spending. Small Business. Cici, E. N., and Bilginer Özsaatcı, F. G. (2021). The impact of crisis perception on consumer purchasing behaviors during the COVID-19 (coronavirus) period: a research on consumers in Turkey. Eski¸sehir Osmangazi Üniversitesi Ýktisadi ve Ýdari Bilimler Dergisi 16, 727–754. doi: 10.17153/oguiibf.923025 Das, S. (2022). The danger of not researching your consumers' decision-making. LinkedIn. https://www.linkedin.com/pulse/danger-researching-your-consumers-decision-making-soumitri-das/?trk=pulse-article\_more-articles\_related-content- card Değermen, A. & Mohammadabbasi, M. (2023). Using big data in analysis of consumer behavior: A qualitative study. Kırklareli Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi. https://doi.org/10.53306/klujfeas.1220342 Indeed. (2022). FAQ: What is consumer spending? (and why it's important). Sauro J, Lewis J. Standardized Usability Questionnaires in Quantifying the User Experience: Practical Statistics for User Research. Waltham, MA 02451: Elsevier; 2012. pp. 193–198. Orlovskyi, D., & Kopp, A. (2020a). A business intelligence dashboard design approach to improve data analytics and decision making Rosário, A., & Raimundo, R. (2021). Consumer marketing strategy and e-commerce in the last decade: A literature review. In Journal of Theoretical and Applied Electronic Commerce Research (Vol. 16, Issue 7, pp. 3003–3024). MDPI. Siti, R. H. A., Bismo, A., & Sutiyo, L. (2019). Segmentation analysis of Instagram users based on preferences towards forms and types of online marketing content. Tao, H., Sun, X., Liu, X., Tian, J., & Zhang, D. (2022). The impact of consumer purchase behavior changes on the business model design of consumer services companies over the course of COVID-19. Frontiers in Psychology, 13. https://doi.org/10.3389/fpsyg.2022.818845 ThinkSecure Network. (2021). 5 ways bad customer service affects your business. [https://www.thinksecurenet.com/blog/5-ways-bad-customer-service-affects- your-business/](https://www.thinksecurenet.com/blog/5-ways-bad-customer-service-affects-%20your-business/) Xeo Blog. (2022). 4 problems you may face without a business dashboard. https://www.xeosoftware.com/4-problems-you-may-face-without-a-business- dashboard/

A picture containing text, clipart Description automatically generated

 © 2024 by the authors. Submitted for possible open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
1. \* Corresponding author. *E-mail address*: (Add the e-mail in the final camera-ready submission)