E-commerce is a key move that has altered the way businesses are conducted, particularly in the retail industry. In the e-commerce business, pricing is one of the most significant aspects in determining profitability, and closely linked to the company's sales. The rise of online markets has necessitated the creation of a Machine Learning tool for pricing suggestions. The aim of this study is to select a machine learning model to create a price suggestion tool for Ecommerce enterprises. Three machine learning algorithms – Linear Regression, Random Forest and LightGBM – are tested on a dataset of an Ecommerce enterprise to indicate the performance of models when using in a dataset with several features and millions of rows. The study also processes to make improvement the output of models, including parameters tuning, feature selection or remove outstanding values. The result shows that LightGBM after GridsearchCV process outperforms in terms of both prediction error and processing time.