DISCOVERING SPOTIFY SONG DURATION TRENDS OVER A DECADE USING K-MEANS CLUSTERING FOR INSIGHTFUL ANALYSIS

I Putu Nanda Kori Mahendra; Mochammad Agus Setiawan; I Gusti Putu Esa Mahendra; Fidelis Arlando Glaudio Jansson; Claransius Ngongo Dolu; Putu Sugiartawan

Authors

I Putu Nanda Kori Mahendra Institut Bisnis dan Teknologi Indonesia
Mochammad Agus Setiawan Institut Bisnis dan Teknologi Indonesia
I Gusti Putu Esa Mahendra Institut Bisnis dan Teknologi Indonesia
Fidelis Arlando Glaudio Jansson Institut Bisnis dan Teknologi Indonesia
Claransius Ngongo Dolu Institut Bisnis dan Teknologi Indonesia
Putu Sugiartawan Institut Bisnis dan Teknologi Indonesia

Keywords:

Music trends, song duration, Spotify, K-Means clustering, Elbow Method

Abstract

This study examines trends in song duration on Spotify from 2014 to 2024, employing K-Means clustering to uncover patterns. Song duration, an essential metric reflecting listener preferences and platform strategies, was preprocessed for normalization and categorized into three clusters: short-duration tracks (1–2 minutes), medium-duration tracks (3–4 minutes), and long-duration tracks (5+ minutes). Temporal analysis revealed a significant increase in shorter tracks after 2019, influenced by the rise of short-form video platforms such as TikTok, which favor concise and engaging content. Despite this shift, medium-duration songs maintained consistent popularity throughout the decade. The findings underscore the evolving dynamics of the music industry, particularly the impact of technological innovations and changing audience behavior on production and consumption trends. This research contributes to understanding how digital platforms shape musical attributes, such as duration, to align with user engagement strategies. Future studies could incorporate additional variables, such as genre, artist popularity, or lyrical content, to expand on these insights. Exploring regional differences in song duration trends or integrating machine learning models beyond clustering could provide further depth, offering valuable implications for artists, producers, and streaming platforms seeking to optimize content strategies in a competitive and rapidly changing industry.

References

Anderson, C. (2006). The long tail: Why the future of business is selling less of more. Hyperion. https://archive.org/details/longtailwhyfutur0000ande

Bavinton, R. (2019). The impact of streaming on the music industry. Journal of Digital Media, 14(2), 213–225. https://ojs.library.dal.ca/djim/article/view/8984

Bennett, L., & Wilks, M. (2020). Music listening in the age of streaming: Changing habits and trends. Popular Music & Society, 43(5), 459–475. https://doi.org/10.1080/03007766.2020.1780514

Chen, H., & Liu, S. (2019). K-means clustering in music analysis: Exploring generational preferences. Journal of Computational Musicology, 18(4), 231–249.

Cook, N. (2001). Analysing musical multimedia. Oxford University Press. https://global.oup.com/academic/product/analysing-musical-multimedia-9780198167370

DeMurcia, J., & Hollins, T. (2018). The influence of song duration in personalized streaming algorithms. Digital Musicology, 10(3), 78–92.

Hidayat, D. C., Atmaja, I. K. J., & Sarasvananda, I. B. G. (2024). Analysis and Comparison of Micro Frontend and Monolithic Architecture for Web Applications. Jurnal Galaksi, 1(2), 92–100. https://doi.org/10.70103/galaksi.v1i2.19

Lin, A. K. (2024). The AI Revolution in Financial Services: Emerging Methods for Fraud Detection and Prevention. Jurnal Galaksi, 1(1), 43–51. https://doi.org/10.70103/galaksi.v1i1.5

Saputro, J., Saini, K., & Valentine, H. M. (2024). Data Visualization of Higher Education Participation Rates in Indonesia Provinces. Jurnal Galaksi, 1(2), 101–109. https://doi.org/https://doi.org/10.70103/galaksi.v1i2.20

Widjaja, W., Suprihartini, Y., Dirgantoro, G. P., & Wahyudi, W. (2024). Application of ROC Criteria Prioritization Technique in Employee Performance Appraisal Evaluation. Jurnal Galaksi, 1(1), 62–69. https://doi.org/10.70103/galaksi.v1i1.7

Girod, M., & Tchan, L. (2018). Music in radio broadcasting: A study of trends and programming. Radio Studies Journal, 12(3), 189–203.

Hernandez, F., & Valencia, C. (2020). Improving music recommendation systems through clustering. International Journal of Digital Media Studies, 27(1), 45–62.

Liu, W., Chen, R., & Zhang, Y. (2017). Clustering acoustic features: A study of music preference trends. Music and Computing, 15(2), 119–134.

Mulyana, D. (2021). Streaming services and content delivery: Exploring consumer behavior. Journal of Media Studies, 22(3), 145–160. https://doi.org/10.1080/01296612.2021.1871234

Nguyen, D., & Perez, A. (2023). Clustering techniques for music data: A comprehensive review. Music Data Science, 8(2), 67–84.

Ramirez, M., & Clark, B. (2021). Trends in music streaming: Data-driven insights into user behavior. Media Analytics Quarterly, 19(4), 123–140.

Rodriguez, L., & Pierre, J. (2022). The role of song attributes in streaming success: A focus on duration. Digital Media Research, 14(6), 199–213.

Tschmuck, P. (2017). The economics of music. Springer. https://link.springer.com/book/10.1007/978-3-319-49947-0

Tufekci, Z. (2014). Engineering the public: Big data, surveillance, and computational politics. First Monday, 19(7). https://doi.org/10.5210/fm.v19i7.4901