This study proposes an efficient and ethically grounded methodology for extracting large-scale product review data from Amazon, aimed at supporting data-driven research in e-commerce analytics. Unlike conventional scraping practices, the novelty of this work lies in integrating a technically rigorous pipeline with explicit ethical safeguards, making it both reproducible and adaptable to diverse research contexts. The approach encompasses environment setup, URL and DOM structure analysis, HTTP request management, and structured data storage using Python libraries such as Requests, BeautifulSoup, and Pandas. Particular attention is given to the practical challenges of DOM-based scraping, including frequent HTML structure changes, nested elements, pagination, and dynamically loaded content, which often hinder consistent data extraction in e-commerce platforms. Ethical scraping practices are embedded throughout the pipeline, including User-Agent rotation, rate limiting, and compliance with website terms of service, thereby minimizing detection risks and ensuring responsible data use. The resulting dataset is structured to enable downstream applications such as sentiment analysis, customer feedback modeling, and recommendation system development. By addressing both the technical complexities of DOM-based scraping and the ethical considerations of large-scale data collection, this case study contributes a novel, scalable, and responsible framework for e-commerce research.