Instagram Reels Engagement Drivers: Machine Learning Insights into Celebrities and Brands
MANAS Sosyal Araştırmalar Dergisi · 26.07.2026 · 15 (3) · 1075-1094
DOI: 10.33206/mjss.1879442 ↗Kurz gefasst
Fachartikel in der Zeitschrift MANAS Sosyal Araştırmalar Dergisi, online erschienen am 26.07.2026. Verfasst von Elif Hasret Kumcu und 1 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben.
Abstract (Originalsprache)
The paper explores the major motivators of user interaction on Instagram Reels, a rising short video form in digital marketing. Despite the central role played by Reels in content strategies, there is still minimal empirical research on which specific features are relevant in driving the engagement. To fill this gap, we analyzed 438 Instagram Reels of global celebrities and brands in the period from January to March 2024. A combination of exploratory data analysis with a Random Forest regression model was applied in a specific way so that it could estimate the importance of the features and find the most significant predictors of the engagement rate. These were the ranked features, which then were the input of a K-Means clustering analysis which indicated patterns of content that were linked to different degrees of audience response. The results indicate that celebrity and brand presence is the most influential factor (22.10% importance), followed by content type (19.03%), age representation (16.27%), and human activity (16.16%). In contrast to common assumptions, the findings show that neutral emotional tone generates the highest engagement, while positive tone yields the lowest, suggesting that authenticity and subtle emotional expression may be more effective than overt positivity in short-form video contexts. The combination of machine learning methods and traditional engagement theory within the study gives practical recommendations to the marketer and will create a clear, reproducible map of research on short-form video analytics in the future.
Zitieren
APA
Kumcu, E. H., & Özçifçi, V. (2026). Instagram Reels Engagement Drivers: Machine Learning Insights into Celebrities and Brands. MANAS Sosyal Araştırmalar Dergisi, 15(3), 1075-1094. https://doi.org/10.33206/mjss.1879442
Vancouver
Kumcu EH, Özçifçi V. Instagram Reels Engagement Drivers: Machine Learning Insights into Celebrities and Brands. MANAS Sosyal Araştırmalar Dergisi. 2026;15(3):1075-1094. doi:10.33206/mjss.1879442
BibTeX
@article{Kumcu2026,
author = {Kumcu, Elif Hasret and Özçifçi, Vesile},
title = {Instagram Reels Engagement Drivers: Machine Learning Insights into Celebrities and Brands},
journal = {MANAS Sosyal Araştırmalar Dergisi},
year = {2026},
volume = {15},
number = {3},
pages = {1075--1094},
doi = {10.33206/mjss.1879442}
}
EUVIMED