Analýza a ověření efektivních strategií pro tvorbu úspěšného YouTube videa (Případová studie)

This bachelor thesis focuses on the analysis and optimization of video content creation strategies on the YouTube platform within the context of modern recommendation systems. The main objective of the thesis is to identify success factors influencing the platform's neural networks and subsequently verify them empirically through a case study conducted on the author's channel, "Banda Chlupatých Opic". The theoretical part critically analyzes the evolution of algorithms from the View Count metric to current deep learning-based models. It compares official claims about the importance of Watch Time with independent studies that identify the Click-Through Rate (CTR) as a critical initial metric for smaller channels. The practical part is designed as a pre-test/post-test experimental study. It compares the performance of a control sample (original content) with an experimental sample of videos where strategies regarding metadata optimization visual psychology (thumbnails), and dramaturgy (audience retention) were consistently applied. The obtained data are evaluated using the YouTube Analytics tool. The output of the thesis is a set of data-driven recommendations for aspiring creators, reflecting the current mechanisms of algorithmic content distribution.