When it comes to the fast-moving world of e-commerce, one of the elements that cannot be abolished regarding the improvement of customer satisfaction and in boosting the financial outcomes is personalized product suggestions. Nevertheless, quite contrary to the outstanding achievements of large-scale corporations like Amazon and Flipkart, most of the existing recommendation systems utilize only purchase history as a source of information. The highly focused methodological scope is bound to result in less attractive recommendations which are often lowly rated or held by little opinions hence triggering the capacity to attract customer dissatisfaction and subsequent loss of revenue. To overcome this deficiency, we present a new recommendation architecture that amalgamates quantitative data of purchase history with product review and ratings leveraging Amazon dataset and Flipkart datasets, respectively. The combination of these data sources leads to more effective recommendations, contributing to the increased rate of customer satisfaction and possibly to trust in the very platform on which customer builds these data. The discussion below presents the shortcomings of the conventional recommendation strategies and introduces a technologically as well as data-intensive solution that enables e-commerce customers with better fitted product choices.