Google Professional Data Engineer Question 176
Select 3Google Cloud PlatformYou are working as a data engineer for a retail company that wants to visualize sales data in Looker Studio. The sales data is stored in BigQuery and consists of multiple tables, including 'transactions', 'products', and 'customers'. To prepare the data for visualization, you need to ensure the data model is optimized and easy to understand for the analysts. Which of the following actions should you take?
- A
Create a single denormalized table by joining the 'transactions', 'products', and 'customers' tables to simplify the data model.
- B
Use appropriate naming conventions and add descriptive column names to make the schema user-friendly.
- C
Ensure that all columns in the joined tables are included in the final dataset for maximum flexibility.
- D
Test the data quality by validating key metrics and checking for missing or inconsistent data.
- E
Apply filtering and aggregation in the visualization tool rather than in the query to maintain raw data availability.
Show answer and explanation
Correct answers: A, B, D
Explanation
To prepare data for visualization, you should focus on simplifying the data model, ensuring data quality, and making the schema user-friendly. Actions like denormalizing tables, using clear naming conventions, and validating data quality help create a robust foundation for accurate and efficient visualizations. Including unnecessary columns or relying on the visualization tool for filtering/aggregation can negatively impact performance and usability.
- A. Correct.
Creating a single denormalized table is a best practice for visualization purposes as it reduces complexity and improves query performance.
- B. Correct.
Using clear naming conventions and descriptive column names makes the data model easier for analysts to understand and use effectively.
- C. Incorrect.
Including all columns in the final dataset can lead to unnecessary complexity and larger query costs, which is not advisable for visualization purposes.
- D. Correct.
Testing data quality ensures that the visualizations are based on accurate and consistent data, which is critical for decision-making.
- E. Incorrect.
Applying filtering and aggregation in the visualization tool can lead to inefficient queries and performance issues; it is better to handle these in the query or data preparation step.