Visualizing Content Performance Analytics for Strategic Decision-Making in News Portal

Visualizing Content Performance Analytics for Strategic Decision-Making in News Portal

Muhamad Dody Firmansyah, Zurina Saaya^*

Faculty Technology Maklumat and Komunikasi, University Technical Malaysia Melaka

*Corresponding Author

DOI: https://dx.doi.org/10.47772/IJRISS.2025.906000174

Received: 29 May 2025; Accepted: 03 June 2025; Published: 05 July 2025

ABSTRACT

In the digital era, news portals have become essential platforms for delivering timely news and information across diverse topics such as politics, business, technology, and public affairs. These platforms, which may be operated by independent media, private corporations, or government institutions. It also generates extensive user engagement data including metrics like page views, reading duration, and user interactions. Despite the abundance of this data, many editorial teams lack the analytical tools necessary to extract actionable insights to guide content strategy and audience engagement. This study explores the application of Tableau, a data visualization tool, to interpret and present user interaction data from the news portal of the Indonesian National Police. As a public sector platform, the portal plays a strategic role in government communication, transparency, and citizen engagement. Through the development of interactive dashboards and visual analytics, this research aims to support data-driven editorial decision-making by providing intuitive insights into content performance and user behaviour. Data extracted from SQL-based systems was processed using Tableau to create visualizations such as bar charts, pie charts, and trend lines depicting content uploads by category, type, and region over time. These elements were integrated into an interactive dashboard that offers editorial teams intuitive insights into content trends. To evaluate system usability, the study employed the System Usability Scale (SUS), a ten-item questionnaire used to assess the dashboard’s effectiveness and ease of use. The final dashboard provided actionable recommendations to enhance content strategy and resource allocation across different work units and regions. This research highlights how visual analytics can significantly improve public sector communication and editorial planning through user-centric, data-driven approaches.

Keywords: news portal, visualization, data analytics, decision-making

INTRODUCTION

A news portal is an online platform that provides timely access to news and information on a wide range of topics, including politics, business, technology, sports, and entertainment. These platforms typically feature headlines, journalist-authored articles, multimedia content such as videos and images, as well as interactive elements like comment sections and social sharing features. Prominent examples include global news portals such as BBC News and CNN, as well as various regional and local online newspapers. News portals may operate independently, be affiliated with media organizations, or be run by government entities. In the digital age, news portals generate a vast amount of user interaction and engagement data for each published news item. This includes metrics such as page views, click-through rates, time spent on articles, and user feedback. Such data is a valuable asset that can provide deep insights into audience behaviour, content performance, and user preferences.

Despite the availability of rich usage data, many news organizations lack effective tools or strategies to systematically analyze and visualize this information. As a result, editorial teams may struggle to make data-driven decisions regarding content curation, headline optimization, publishing schedules, and audience engagement strategies. The absence of a clear, visual representation of news consumption patterns hinders the ability to plan and adapt editorial directions based on audience needs and preferences. This study aims to explore the use of Tableau, a data visualization tool, to analyze and visually present usage data from a news portal. The objective is to demonstrate how interactive dashboards and visual analytics can support editorial teams in interpreting user engagement data more effectively, thereby enabling more informed editorial planning and content strategy development. This study will focus on usage data collected from the news portal operated by the Indonesian National Police and can be accessed at mediahub.polri.go.id. This portal serves as a public communication platform to disseminate up-to-date news and information about the department’s activities, events, and initiatives within the region. As a government-operated media outlet, it plays a critical role in fostering transparency, building public trust, and engaging with citizens through timely and accurate reporting.

RESEARCH BACKGROUND

User Engagement Data in News Portal

The rise of digital media has transformed how news is produced, distributed, and consumed. News portals, as digital platforms, generate significant volumes of usage data that can inform editorial and strategic decisions (Thurman & Schifferes, 2012). Metrics such as click-through rates, bounce rates, page views, and session duration provide measurable indicators of user engagement and content effectiveness (Cherubini & Nielsen, 2016). However, many organizations lack structured processes or tools for turning raw data into actionable insights (Peters, 2011). In institutional or government-operated news platforms, the goal is often twofold: to inform the public and to enhance institutional transparency. Understanding which types of content garner attention can help these platforms fulfill their public service missions more effectively (Mak & Song, 2019).

Digital Communication Strategies in Government Institutions

Government institutions face unique challenges in public communication, including the need for accuracy, credibility, and public trust. Digital platforms have become vital tools for disseminating information, especially in contexts involving law enforcement, health, or disaster response (Mergel, 2013). Studies show that well-maintained government news portals can enhance transparency, civic engagement, and institutional legitimacy (Criado et al., 2013). However, there is limited research on how government communication teams use analytics to inform content decisions. This gap underscores the importance of exploring how visualization tools can support public sector communication strategies, particularly in non-commercial or authoritative settings like that of the Indonesian National Police.

Data Visualization

Data forms the foundational component of data visualization, yet its nature and role require careful consideration. Scholars widely describe data as raw, unprocessed information that lacks inherent meaning unless analyzed or structured (Kitchin, 2014). Data can take various forms, including numerical values, symbols, textual content, or even unstructured signals, and serves as a digital representation of real-world or simulated attributes (Borgman, 2015). Crucially, data in its raw state is neutral—it exists independently of interpretation, and its significance only emerges when contextualized through systematic processing (Helmond, 2014; John Walker, 2014). Understanding this transformation from raw data to meaningful insight is essential in designing effective visualizations that support analysis and decision-making.

Data visualization is the practice of converting processed data into graphical representations to enhance comprehension, analysis, and decision-making (Few, 2009; Kirk, 2016). Researchers emphasize its role as a critical tool in business intelligence, where complex datasets are rendered interpretable through visual means (Wexler et al., 2017). Among the most widely used visualizations are bar charts, which represent categorical data through vertical or horizontal bars whose lengths correspond to quantitative values, enabling straightforward comparisons across categories or time periods (Evergreen, 2014). Line charts, on the other hand, are particularly effective for tracking trends over time, as they connect data points with continuous lines to highlight patterns and fluctuations (Knaflic, 2015).

Pie charts offer a circular depiction of part-to-whole relationships, with each slice representing a proportion of the total. However, their efficacy is highest when applied to datasets with a limited number of categories to preserve clarity (Cairo, 2016). Scatter plots provide a powerful means of visualizing the relationship between two variables by positioning data points along Cartesian coordinates, thereby revealing correlations, clusters, or outliers. These can be further enriched through the use of colour, size, and shape to introduce multidimensional insights (Tufte, 2001). Lastly, histograms present frequency distributions of continuous data by organizing values into contiguous intervals or bins, making them particularly useful for examining the spread, central tendency, and shape of data distributions (Few, 2009). Collectively, these visualization techniques serve as critical tools in transforming raw datasets into actionable insights.

Data visualization has become an essential tool in modern journalism, not only for storytelling but also for internal analysis. Visualization tools like Tableau enable users to transform complex datasets into interactive, interpretable formats that support decision-making (Kirk, 2016). Research shows that visualization enhances cognitive understanding and pattern recognition, making it a valuable method for analysing trends in news consumption (Cairo, 2016; Few, 2009). Moreover, newsrooms and communication teams increasingly rely on dashboards to monitor real-time performance and audience trends (Anderson, 2013). These tools help bridge the gap between data and editorial strategy by providing timely, actionable insights.

Data Visualization Tools

Several powerful tools exist to facilitate data visualization and analysis, each offering unique capabilities suited for different analytical needs. Among these, Tableau stands out as a user-friendly business intelligence platform that simplifies data visualization, analysis, and reporting. What makes Tableau particularly valuable is its ability to integrate data from multiple sources—including spreadsheets, databases, cloud storage, and big data repositories—into a single unified interface. Furthermore, its intuitive drag-and-drop functionality enables users to perform dynamic analysis without requiring extensive technical expertise (Murray, 2016).

Another prominent solution in this area is Microsoft Power BI, a versatile analytics platform that supports diverse data sources such as SQL databases, Excel files, open-source datasets, and web-based data. Notably, Power BI offers real-time data visualization capabilities through its web-based interface, making it both flexible and widely accessible. This accessibility and functionality have earned Power BI recognition as one of the top applications for business intelligence development (Microsoft, 2023; Czerwinski & others, 2020).

For more advanced analytical needs, Python has emerged as the programming language of choice for data professionals. As a high-level, object-oriented language, Python excels at processing large and complex datasets. Its popularity among data scientists stems from its remarkable flexibility, supporting everything from data modelling and systematization to developing machine learning algorithms and web services. Particularly relevant for visualization, Python offers an extensive ecosystem of specialized visualization packages that enable the creation of sophisticated charts, graphs, and interactive visual interfaces (McKinney, 2022; VanderPlas, 2016). This combination of features makes Python an indispensable tool for modern data analysis and data science workflows (Cao, 2017). While each tool has its strengths, they all share the common goal of transforming raw data into actionable insights through effective visualization techniques. The choice between them typically depends on the specific requirements of the analysis, the technical expertise of the users, and the scale of the data being processed.

METHODOLOGY

This study employs a data-driven case study approach to explore how Tableau can be used to visualize and interpret user engagement data from the Indonesian National Police’s news portal. The methodology consists of five main stages: problem definition, dataset construction, report generation, dashboard development, and evaluation, refer to Figure 1.

Figure 1: Methodology

The first stage, problem definition, involves identifying the lack of effective tools for visualizing and interpreting user interaction data within the editorial process. Despite the availability of large volumes of engagement data such as page views, visit durations, and regional access patterns. There remains a gap in how this data is utilized to guide editorial strategy. The aim at this stage is to clarify the need for a visualization solution that can support informed decision-making by news editors (Cooper et al., 2010).

In the second stage, build dataset, the research follows the Extract, Transform, Load (ETL) process to prepare the data for visualization. Extraction involves retrieving user interaction data directly from the news portal’s backend database, which is managed using PostgreSQL. This relational database contains structured data related to article metadata, activity logs, user comments, timestamps and many more. Transformation includes cleaning and structuring the data, removing inconsistencies, standardizing formats, and aggregating metrics such as article views, session duration, user location, and publication timestamps. Loading refers to importing the refined data into a structured format suitable for visualization, such as CSV or Excel files, which are then connected to Tableau for further analysis. Tools like Microsoft Excel, Python (for scripting), or SQL may be used during this stage to handle the data transformation effectively(Kelleher & Tierney, 2018).

The third stage involves the generation of a descriptive statistical report that summarizes trends across the dataset. This report includes basic visual and tabular summaries of article performance, user engagement by category, traffic over time, and geographical access patterns. The purpose of this stage is to offer an initial understanding of the data and to highlight areas that may warrant further attention during the dashboard development phase (Provost & Fawcett, 2013).

In the fourth stage, dashboard development, the structured dataset is imported into Tableau to create interactive and dynamic visualizations. The dashboard features multiple elements including time-series charts for analyzing traffic trends, bar charts to compare content categories, and heat maps to visualize geographic distribution of users. These visualizations allow editorial staff to interact with the data, apply filters, and gain real-time insights. Tableau’s drag-and-drop interface and visualization flexibility make it an ideal tool for this purpose (Kirk, 2016).

The final stage, evaluation, assesses the dashboard’s usability, clarity, and practical value. This involves heuristic evaluations, usability testing with scenario-based tasks, and informal feedback sessions with editorial team members. Additionally, this study applies the System Usability Scale (SUS) to quantitatively measure the perceived usability of the dashboard. The SUS is a widely adopted questionnaire consisting of ten statements rated on a five-point Likert scale ranging from “strongly disagree” to “strongly agree.” It is designed to capture the user’s subjective assessment of the system’s usability, efficiency, and ease of use, refer to Table 1. The SUS score provides a standardized measure that facilitates comparison and benchmarking of system usability (Nielsen, 1994).

Table 1: System Usability Scale

Through these five stages, the research aims to demonstrate the potential of visual analytics in improving public-sector news communication and data-informed editorial planning.

RESULTS AND DISCUSSION

This section describes the architectural design in implementing the visualization dashboard with the concept of a user-friendly prototype using Tableau.

Description Analytic

The bar chart in Figure 2 visualizes data on the number of content uploads based on work units in a dataset that includes photo, text, audio and audio-visual content.

Figure 2: Predictive Result (Analysis Content)

The purpose of the image is to visualize the raw data and facilitate interpretation between users. Then the raw data will be analysed using the Knowledge Data Management (KDM) technique. The KDM process is carried out in four stages; data selection, data pre-processing, data transformation, and data interpretation. In this analysis, pie and bar charts are used to visualize the percentage of each category. This section contains a pie chart to visualize the total usage that uploads content, based on work units and regions based on audio, visual and photo content categories. and the last step at the bottom of the dashboard displays a graph showing the total users uploading content in a time range and month.

Dashboard Predictive Analytic

After successfully developing a series of visualizations including a bar chart of top-uploaded content categories by type, a bar chart of total uploads across content categories and types, and a trend line chart illustrating content upload trends over time the visual components were finalized. These elements were then integrated into an interactive and visually engaging dashboard as depicted in Figure 3. This dashboard aims to facilitate a more intuitive and effective understanding of content upload patterns on Indonesian news portals, offering editorial teams’ actionable insights at a glance.

Figure 3: Dashboard Predictive (Analysis Content)

To support the formulation of recommendations based on user content-upload activity, prescriptive analysis was conducted using the System Usability Scale (SUS) framework. The SUS, a well-established tool for evaluating system usability, was applied to assess the dashboard’s effectiveness from the user’s perspective. Additionally, predictive models were employed to identify and forecast the most active content upload categories. These models were evaluated based on their explanatory power the extent to which the model can account for variations in the output using the training dataset. A higher explanatory value indicates a more reliable and effective model.

The prediction model developed in this study focuses on identifying the content categories most frequently uploaded by users of the news portal. The results, visualized in Figure 4, provide recommendations based on the predictive analysis outcomes, offering data-informed guidance for future content planning and resource allocation. As shown in Figure 6, the system aims to support decision-makers by providing actionable recommendations to enhance the performance of content uploads across various work units and regions. These recommendations are based on the percentage of active versus passive users, enabling decision-makers to identify areas for improvement and allocate resources more effectively.

Figure 4: Result Report Dashboard Analytic Portal News of Tableau

SUS (System Usability Scale) Testing

At this stage, an evaluation was conducted to assess the usability of each activity within the dashboard, based on the results of a structured questionnaire using the System Usability Scale (SUS) methodology. This evaluation was guided by a standardized scoring formula, which serves as the foundation for interpreting the usability outcomes and identifying areas for improvement. To gather relevant data, the researcher conducted direct interviews with key personnel responsible for managing the news portal specifically, users who interact regularly with the dashboard for content monitoring and decision-making. These interviews provided qualitative insights into the user experience and informed the subsequent quantitative assessment.

Table 2: SUS Testing Results

CONCLUSIONS

This study has demonstrated the application of data visualization techniques using Tableau to analyze user engagement data from a government-operated news portal. By implementing a structured methodology comprising problem identification, dataset construction using PostgreSQL and the ETL process, statistical reporting, dashboard development, and evaluation through usability testing, the research highlights the practical value of visual analytics in supporting editorial decision-making. The integration of the System Usability Scale (SUS) further provided measurable insights into the effectiveness and user-friendliness of the developed dashboard.

The results indicate that interactive visualizations significantly enhance the ability of editorial teams to interpret complex usage data, identify content trends, and align their strategies with audience interests. This is particularly important in public-sector communication, where transparency, clarity, and responsiveness are essential. Overall, this research supports the adoption of visual analytics tools in digital journalism environments and encourages further exploration into scalable, real-time solutions for public information management.

ACKNOWLEDGEMENT

The authors would like to thank the Faculty Technology Maklumat dan Komunikasi (FTMK), University Technical Malaysia Melaka (UTeM), and the Scenter of Advanced Computing Technology (C-ACT) for their incredible support in this project.

REFERENCES

1. Anderson, C. W. (2013). Rebuilding the news: Metropolitan journalism in the digital age. In Rebuilding the News: Metropolitan Journalism in the Digital Age (Vol. 9781439909355).
2. Borgman, C. L. (2015). Big Data, Little Data, No Data: Scholarship in the Networked. MIT Press
3. Cairo, A. (2016). The Truthful Art: Data, Charts, and Maps for Communication. New Riders.
4. Cherubini, F., & Nielsen, R. K. (2016). Editorial Analytics: How News Media are Developing and Using Audience Data and Metrics. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2739328
5. Cooper, A., Reimann, R., & Cronin, D. (2010). About Face: The Essentials of Interaction Design. Wiley.
6. Microsoft Corporation. (2023). What is Power BI?
7. Criado, J. I., Sandoval-Almazan, R., & Gil-Garcia, J. R. (2013). Government innovation through social media. In Government Information Quarterly (Vol. 30, Issue 4). https://doi.org/10.1016/j.giq.2013.10.003
8. Czerwinski, D., & others. (2020). Microsoft Power BI Data Analyst Certification Guide: A comprehensive guide to becoming a proficient Power BI professional. Packet Publishing.
9. Evergreen, S. D. H. (2014). Presenting Data Effectively: Communicating Your Findings for Maximum Impact. SAGE Publications.
10. Few, S. (2009). Now You See It: Simple Visualization Techniques for Quantitative Analysis. Analytics Press.
11. Helmond, A. (2014). Raw data is an oxymoron. Information, Communication & Society, 17(9). https://doi.org/10.1080/1369118x.2014.920042
12. Cao, L. (2017). Data science: A comprehensive overview. ACM Computer Survey 50(43). https://doi.org/10.1145/3076253.
13. John Walker, S. (2014). Big Data: A Revolution That Will Transform How We Live, Work, and Think. International Journal of Advertising, 33(1). https://doi.org/10.2501/ija-33-1-181-183
14. Kelleher, J. D., & Tierney, B. (2018). Data Science. MIT Press.
15. Kirk, A. (2016). Data Visualisation: A Handbook for Data Driven Design. SAGE Publications.
16. Kitchin, R. (2014). The Data Revolution: Big Data, Open Data, Data Infrastructures & Their Consequences. In The Data Revolution: Big Data, Open Data, Data Infrastructures & Their Consequences. https://doi.org/10.4135/9781473909472
17. Knaflic, C. N. (2015). Storytelling with Data: A Data Visualization Guide for Business Professionals. Wiley.
18. Mak, A. K. Y., & Song, A. O. (2019). Revisiting social-mediated crisis communication model: The Lancôme regenerative crisis after the Hong Kong Umbrella Movement. Public Relations Review, 45(4). https://doi.org/10.1016/j.pubrev.2019.101812
19. McKinney, W. (2022). Python for Data Analysis: Data Wrangling with Pandas, NumPy, and Jupyter (3rd ed.). O’Reilly Media.
20. Mergel, I. (2013). Social media in the Public Sector: A guide to Participation, Collaboration, and Transparency in the Networked World. Jossey-Bass, 22(SPE).
21. Murray, D. G. (2016). Tableau Your Data!: Fast and Easy Visual Analysis with Tableau Software. Wiley.
22. Nielsen, J. (1994). Usability Engineering. Morgan Kaufmann.
23. Peters, C. (2011). Emotion aside or emotional side? Crafting an “experience of involvement” in the news. Journalism, 12(3). https://doi.org/10.1177/1464884910388224
24. Provost, F., & Fawcett, T. (2013). Data Science for Business: What You Need to Know About Data Mining and Data-Analytic Thinking. O’Reilly Media.
25. Thurman, N., & Schifferes, S. (2012). The Future of Personalization at News Websites: Lessons from a longitudinal study. Journalism Studies, 13(5–6). https://doi.org/10.1080/1461670X.2012.664341
26. Tufte, E. R. (2001). The Visual Display of Quantitative Information (2nd ed.). Graphics Press.
27. VanderPlas, J. (2016). Python Data Science Handbook: Essential Tools for Working with Data. O’Reilly Media.
28. Wexler, S., Shaffer, J., & Cotgreave, A. (2017). The Big Book of Dashboards: Visualizing Your Data Using Real-World Business Scenarios. Wiley.