50 AI Prompts for Data AnalystsExplore data faster, write better SQL, and tell stories that drive decisions.

Write a dataset summary for exploratory data analysis. Given this dataset [describe schema or paste column names and types], describe: the shape of the data, what each column represents, likely data types, potential issues to investigate, and the business question this data could answer.

Write an outlier detection explanation for a non-technical stakeholder. I found outliers in [column/metric] using [method — IQR, z-score, etc.]. Explain: what an outlier is, why it matters, how I detected it, and what we should do with it.

Write a distribution analysis narrative for [column/variable] in [dataset]. Describe: shape of the distribution (normal, skewed, bimodal), key statistics (mean, median, mode, std dev), and what the distribution implies for analysis or modeling.

Interpret this correlation matrix for a non-technical audience. Highlight: the strongest positive and negative correlations, which relationships are expected vs. surprising, potential multicollinearity concerns, and which correlations warrant further investigation. Matrix: [paste or describe]

Write a missing data strategy for [dataset] where [X%] of [column] values are null. Options to evaluate: deletion, imputation (mean/median/mode), predictive imputation, or flagging as a separate category. Recommend an approach with reasoning.

Create a data dictionary for [dataset/table]. For each column, provide: column name, data type, description, example values, allowed range or categories, and notes on known data quality issues.

Write a column naming convention guide for a [team/organization] using [SQL/Python/Tableau/etc.]. Cover: naming format (snake_case, camelCase), prefixes for dimension vs. measure vs. flag columns, date field conventions, and examples.

Generate 10 initial hypotheses for a dataset about [business domain — e.g., e-commerce orders, app usage, marketing campaigns]. Each hypothesis should be testable, business-relevant, and suggest a specific analysis to run.

Write a data sanity check checklist for [dataset or pipeline output]. Include: row count validation, null checks, duplicate detection, value range checks, referential integrity checks, and trend vs. prior period comparison.

Write a data quality audit prompt I can use with an AI assistant to systematically assess [dataset]. Cover: completeness, accuracy, consistency, timeliness, and validity — with specific SQL or Python checks for each dimension.

Explain why this SQL query is slow and how to optimize it. Identify: missing indexes, inefficient joins, unnecessary subqueries, and better alternatives. Query: [paste query]. Table sizes and execution context: [describe]

Write documentation for a complex SQL query with multiple joins. For each join: what tables are joined, on what key, the cardinality (one-to-one, one-to-many), and why the join is necessary. Query: [paste query]

Write a CTE (Common Table Expression) documentation template. For each CTE in this query, describe: name, purpose, input data, transformation logic, and output. Query: [paste query with CTEs]

Write inline comments for this stored procedure. Comments should explain: the purpose of each logical block, what each key variable does, why specific logic choices were made, and any known limitations. Procedure: [paste code]

Write a query performance summary for this execution plan. Identify: the most expensive operations, where row estimates diverge from actuals, recommended indexes, and the estimated performance improvement if changes are made. Plan: [paste or describe]

Document this data pipeline for a new team member. Cover: data sources, transformation steps, business rules applied, output tables and views, refresh schedule, known issues, and who to contact for questions. Pipeline: [describe]

Review this SQL query for correctness, efficiency, and readability. Check for: logical errors, implicit conversions, N+1 patterns, missing NULLs handling, and naming clarity. Suggest improvements. Query: [paste]

Translate this SQL query into plain English for a business stakeholder. Explain: what data it retrieves, the filters applied, how tables are combined, and what the final output represents. Query: [paste]

Write a schema explanation for this database. For each table: purpose, primary key, key relationships, important fields, and common query patterns. Schema: [paste DDL or describe tables]

Create a query testing checklist for validating a new or modified SQL query before deploying to production. Cover: edge cases to test, boundary conditions, NULL handling, expected vs. actual row counts, and performance baseline.

Help me choose the right chart type for [data scenario — e.g., comparing categories, showing trends over time, displaying distribution, showing part-to-whole]. Explain the options, their tradeoffs, and recommend the best choice for a [audience type] audience.

Write a dashboard design brief for a [team/business unit] performance dashboard. Cover: purpose, primary audience, key metrics to include (with definitions), layout recommendations, refresh frequency, and design principles to follow.

Write an executive summary narrative from this dataset and analysis. Audience: [C-suite / board / non-technical stakeholder]. Format: 3–5 bullets covering the headline finding, key supporting data, implications, and recommended action. Data: [paste key stats]

Write annotation scripts for 3 key data points on this visualization. For each: what the data point is, why it matters, and what action it suggests. Visualization type: [describe]. Key points: [list]

Write a KPI story for [metric — e.g., DAU, churn rate, gross margin] for the [period] business review. Cover: metric definition, current value, trend, vs. target, root cause of change, and what we're doing about it.

Create a presentation deck outline for sharing [analysis or project findings] with [audience — leadership, clients, cross-functional team]. Cover: slide titles, one-sentence objective per slide, key data to include, and talking points.

Write an A/B test result summary for a non-technical audience. Cover: what was tested, primary metric and result, statistical significance, secondary metric impacts, and recommendation. Test: [describe]

Write a funnel analysis narrative for [conversion funnel — e.g., signup to activation, trial to paid, visit to purchase]. Cover: conversion rate at each stage, biggest drop-off point, comparison to benchmark, and top 2 hypotheses for improvement.

Write a cohort analysis explanation for a [business type] stakeholder. Cover: what cohort analysis is, what cohort we analyzed, what the data shows about retention and engagement over time, and what actions it suggests. Data: [describe findings]

Write metric definition documentation for [metric name]. Cover: business definition, calculation formula, data sources, known limitations or edge cases, how it differs from related metrics, and examples of correct vs. incorrect interpretation.

Write a weekly metrics email for [team/business] covering [period]. Include: headline numbers, comparison to prior period and target, 2–3 key insights, any anomalies to flag, and what to watch next week. Metrics: [list]

Write a monthly business review summary for [department/product/business] for [month]. Cover: performance vs. plan, key drivers of variance, wins, misses, and priorities for next month. Format for a 5-minute executive read.

Write an anomaly alert message for [metric] that moved [X%] vs. [benchmark — prior period, target, baseline]. Cover: what happened, magnitude of change, likely causes, what's being investigated, and timeline for a root cause update.

Write a data request clarification template. A stakeholder asked for [vague or incomplete data request]. My response should: restate what I think they're asking, ask 3 clarifying questions to scope the work, and give a realistic timeline.

Write an analysis findings presentation script for [audience — team, leadership, clients]. Cover: opening context, methodology summary, key findings (3–5 insights), implications, recommended actions, and how to handle Q&A.

Write a non-technical explanation framework for [complex analytical concept — e.g., statistical significance, regression, cohort analysis, confidence interval]. Analogies encouraged. Target audience: [business stakeholder type]

Write executive dashboard commentary for [dashboard name] for [period]. Cover: the 3 most important things the data is telling us, one thing we should watch closely, and one recommended action. Use plain language.

Write an SLA reporting template for [data/analytics service or team]. Cover: SLAs tracked, current performance vs. target, incidents this period, root cause of any SLA misses, and remediation actions.

Write a project status update for [analytics project] at [milestone stage]. Cover: work completed, work in progress, blockers, revised timeline (if any), decisions needed from stakeholders, and next milestone.

Write a findings memo for [analysis or project name]. Format: executive summary (3 sentences), background, methodology, key findings (numbered), implications, recommendations, and appendix pointers. Audience: [describe]

Generate 10 data portfolio project ideas for a [skill level — junior/mid/senior] analyst in [domain — e.g., e-commerce, healthcare, finance, marketing]. For each: project description, dataset source, analysis type, deliverable format, and what skill it demonstrates.

Write a technical blog post outline for [data topic — e.g., a SQL optimization trick, EDA approach, visualization technique]. Cover: hook, problem statement, solution walkthrough, code examples, key takeaway, and CTA for readers.

Write 10 data analyst interview prep questions for a [role level] role at a [company type — startup, enterprise, consulting firm]. For each: the question, what the interviewer is assessing, and a framework for a strong answer.

Write a GitHub README template for a data analytics project. Cover: project title, problem statement, dataset description, methodology, key findings, how to run the code, technologies used, and contact and license info.

Write a case study writeup for [analysis or project]. Format: business problem, data sources, methodology, key findings (with visuals described), business impact, and lessons learned. Audience: [portfolio/interview/client]

Write a data challenge solution walkthrough for [challenge description — e.g., predict churn, analyze cohort retention, build a dashboard]. Cover: approach, assumptions, key steps, results, and what you'd do differently with more time or data.

Write inline comments for this Python or R data analysis script. Comments should explain: what each block does, why specific choices were made (e.g., why this aggregation, why this chart type), and any caveats. Code: [paste]

Optimize my LinkedIn profile for a data analyst role in [industry/specialization]. Rewrite: headline, about section, and top 3 experience bullets. Goal: attract [target — recruiters, hiring managers, collaborators]. Current profile: [paste]

Write a cover letter for a [data analyst / BI engineer / data scientist] role at [company type]. Highlight: relevant skills, a specific project or result, why this company and role, and a confident close. Job description: [paste key requirements]

Write a networking message for connecting with a [data professional — senior analyst, analytics manager, data engineer] at [target company or in target niche]. Context: [cold outreach / mutual connection / conference / online community]. Keep it specific and genuine.