Instagram Reels Engagement Drivers: Machine Learning Insights into Celebrities, and Brands
(A2025-126090)

AUTHORS

Elif Kumcu, Aksaray University

ABSTRACT

This study explores the key factors influencing engagement rates on Instagram Reels, focusing on content attributes, emotional tone, and demographic features. Using data from 438 Reels posted between January 1, 2024, and March 31, 2024, from globally renowned celebrities (e.g., Cristiano Ronaldo, Kylie Jenner) and top brands (e.g., Nike, National Geographic), a mixed-method approach was applied. Techniques included exploratory data analysis, Random Forest regression, and K-means clustering. Results show that content featuring celebrities, creative activities, and human-centric themes drives higher engagement rates. Videos with diverse themes, such as human and animal interactions, outperform less varied content. Key attributes such as activity type, emotional tone, and auxiliary features like sound or writing significantly impact engagement. These findings provide actionable insights for optimizing social media strategies and bridge the gap between traditional statistical methods and advanced machine learning approaches. Keywords: Instagram Reels, Social Media Marketing, Engagement Rate