SQL Report Builder

A clean, commented SQL report query with CTEs, performance notes, and a plain-English walkthrough of the logic.

Best For

•A data analyst turning a stakeholder's plain-English report request into a commented, CTE-structured SQL query ready to run in Snowflake
•An engineer documenting a recurring revenue-by-region report so any teammate can read and modify the query later
•An analytics consultant flagging ambiguous requirements in the schema before writing a query that might return the wrong numbers

Prompt Template

Write a SQL query (or set of queries) to produce the following report: [REPORT_DESCRIPTION]. Database: [DATABASE_TYPE, e.g., PostgreSQL, BigQuery, Snowflake, MySQL]. Available tables and their relevant columns: [TABLE_SCHEMA_DESCRIPTION]. Report requirements: (1) Metrics to calculate: [METRICS_LIST], (2) Dimensions to group by: [DIMENSIONS], (3) Filters to apply: [FILTERS, e.g., date range, active users only, specific regions], (4) Sort order: [SORT_ORDER], (5) Output format: [OUTPUT_FORMAT, e.g., one row per day per segment / one summary row per user]. Write the query with: (a) clear CTE (Common Table Expression) structure for readability, break complex logic into named steps, (b) inline comments on any non-obvious logic, (c) a note on expected query performance and any indexes that should exist for efficiency. After the query, provide a plain-English explanation of what each CTE or major section does. If any requirement is ambiguous or the schema appears to be missing data, flag it with a [QUESTION: ...] comment in the code.

Pro Tip

Require the model to flag ambiguous requirements with inline [QUESTION:] comments rather than guessing, since silently assumed filters are the most common source of wrong numbers in stakeholder reports.

Example Output

A sample of what this prompt produces once you fill in the placeholders.

WITH monthly_orders AS (SELECT date_trunc('month', order_date) AS month, region, SUM(total) AS revenue FROM orders WHERE status = 'completed' GROUP BY 1, 2) SELECT month, region, revenue FROM monthly_orders ORDER BY month DESC, revenue DESC. Plain-English note: the CTE aggregates completed orders by month and region before the final query sorts by most recent month and highest revenue.