The exponential increase in social media and video-sharing websites, fueled by Web 2.0 technologies, has resulted in a surge in user-generated content. Of all the social media websites, YouTube is one with the most users. This study concerns the matter of classifying the YouTube video titles as clickbait or non-clickbait. Previous works utilized different machine learning algorithms for the same purpose but, in most cases, with little exploratory data analysis (EDA) and without utilizing advanced techniques such as BERT classification. In this research study, we perform extensive EDA, preprocessed the data, and used machine learning algorithms such as Multinomial Naive Bayes, Support Vector Machine (SVM), Random Forest, and Bidirectional Encoder Representations from Transformers (BERT). Our aim is not only to classify the clickbait titles correctly but also enhance user experience, increase advertiser trust, and enhance YouTube’s recommendation system. Through this endeavor, we provide valuable contribution towards the detection of clickbait, adding to the shaping of the dynamic nature of content moderation and user engagement on social media websites.