A data analyst is tasked with understanding customer

Solution

Data cleaning is an essential first step after collecting raw data, ensuring the dataset is accurate, consistent, and usable. Cleaning involves handling missing values, removing duplicates, correcting inaccuracies, and standardizing formats. For example, in a customer churn analysis, incomplete demographic information, inconsistent subscription statuses, or duplicate entries could skew results. By addressing these issues upfront, the data analyst lays a solid foundation for reliable analysis, avoiding errors in downstream processes such as EDA, modeling, or visualization. Cleaning ensures data integrity, which is critical for building models or interpreting trends accurately. Why Other Options Are Incorrect: • A: Building predictive models without clean data can lead to flawed or unreliable predictions. • B: EDA should follow data cleaning to ensure the trends and patterns observed are valid. • C: Visualization comes after data analysis and modeling, not before. • D: KPIs should be defined during the planning phase, before collecting and cleaning data.