Today's multi-billion-dollar online advertising industry is restlessly focused on maximizing return on investment (ROI) and good campaign strategies derive after gaining insight from user and advertisement data definitely helps to achieve good ROI. In this research we explore the usage and implementation of Exploratory Data Analysis (EDA) to gain insight for Online advertisement dataset to enhance the prediction of CTR (Click Through Rate) using machine learning (ML) techniques. Dataset collected from Mendely contains attributes regarding user demographics like Gender, Age, Income, Location with Advertisement related attributes like Type, Topic, Advertisement placement etc. With the analysis goal to uncover insights into user engagement patterns and ad performance metrics which helpful to build better strategic advertisement campaign. With the key steps, EDA start with data cleaning and pre-processing, univariate analysis, bivariate analysis, and end with multivariate analysis. Initial data cleaning includes handling missing values, detecting and addressing duplicates, identifying outliers and encoding for categorical features. Using Univariate analysis visualization techniques distribution of individual variables are examined. The output reveal the insight pattern of various categorical variable interaction and combination of multiple categorical and numerical variable. The knowledge gained from the EDA helpful for refining ML models to predict CTR aimed to designing effectiveness and optimizing advertising strategies to target customer and generate better ROI.