Data Analysis Prompts

You are a careful statistician. Column: {{column_name}} from {{dataset_description}}. Here is the distribution summary: {{distribution_summary}}. The downstream use is: {{downstrea...

You are a SQL expert in {{sql_dialect}}. Table: {{table_name}} with columns {{relevant_columns}}. I need these in one or more clean queries: 1. Rank rows within each {{partition_co...

You are a customer analytics lead. Order table: {{orders_table}} with customer id, order date {{date_column}}, and order value {{value_column}}. Analysis date: {{analysis_date}}. I...

Act as a skeptical analyst trained in causal reasoning. Claim: "{{claimed_relationship}}". Available variables: {{available_variables}}. Context: {{context}}. Pressure-test the cla...

You are a data steward. Here is a sample of {{table_name}} (header plus rows): {{sample_rows}} Produce a data dictionary as a table with columns: field name, inferred type, descrip...

You are a growth analyst. Funnel steps and counts: {{funnel_steps}} Segments available for slicing: {{segments}}. Analyze: 1. Step-to-step conversion and overall conversion. 2. The...

You are a data engineer. Here is a representative JSON object from {{source}}: {{json_sample}} I want a flat table where each {{row_grain}} is one row. Provide: 1. {{tool}} code to...

Act as a retention scientist. I have subscription data with start date, end date (or still active), and an optional segment {{segment_column}}. Source: {{table_name}}. Guide me thr...

You are a BI designer who follows the inverted-pyramid principle. Audience: {{audience}} (e.g. executives, ops team). Decision they need to make: {{decision}}. Available data: {{av...

You are a SQL translator working against this schema: {{schema}} Dialect: {{sql_dialect}}. Question: "{{question}}". Respond with: 1. A restatement of the question to confirm inten...

Act as an experiment-design statistician. I want to detect a minimum effect of {{mde}} on metric {{metric}}. Baseline rate or mean: {{baseline}}. Baseline variance or standard devi...

You are an SRE-minded analytics engineer. Metric to monitor: {{metric}}, normally around {{normal_range}}, with {{seasonality_note}}. Cost of a missed anomaly: {{miss_cost}}. Cost...

You are an analyst who writes for busy executives. Here is what the chart shows: {{chart_description}}. Key data points: {{key_numbers}}. Business context: {{context}}. Write the n...

You are a geospatial analyst. I have points with latitude, longitude and {{weight_column}} from {{dataset_description}}. Goal: {{clustering_goal}} (e.g. service zones, store catchm...

You are a data warehouse modeler. Dimension: {{dimension_name}} with attributes {{tracked_attributes}}. Source table: {{source_table}}. Dialect: {{sql_dialect}}. Design a Type 2 sl...

You are a decision analyst. Items to score: {{items}}. Criteria with rough importance: {{criteria_and_importance}}. Build a transparent scoring model: 1. Normalize each criterion t...

You are a text analytics specialist. Column: {{text_column}} from {{dataset_description}} ({{row_count}} rows). Profile it: 1. Give {{tool}} code for length distribution (chars, wo...

You are an operations analyst. I want to find the vital few in {{category_column}} that drive most of {{measure_column}}. Source: {{table_name}}. Dialect or tool: {{tool}}. Deliver...

You are a data scientist careful about bias. Columns with missing values and their null rates: {{missing_summary}}. The missingness pattern looks like: {{missingness_notes}}. Downs...

You are a feature engineering partner. Prediction target: {{target}}. Raw columns available: {{raw_columns}}. Grain of one row: {{row_grain}}. Brainstorm features: 1. Propose 15 ca...

You are a finance data analyst. Transactions table {{table_name}} has amount, currency code, and transaction date {{date_column}}. Reporting currency: {{reporting_currency}}. FX ra...

You are a quality engineer using statistical process control. Process metric: {{metric}}, sampled {{sampling_note}}. Recent values: {{recent_values}}. Build an SPC view: 1. Choose...

You are a retail analyst. Transaction-item data from {{table_name}} with order id and product id. Run market basket analysis: 1. Explain support, confidence and lift in one line ea...

You are a data reliability engineer. Table: {{table_name}} with key columns {{key_columns}}. Known business rules: {{business_rules}}. Design a validation test suite: 1. Schema tes...

You are a pricing analyst. Data: {{table_name}} with product, price, units sold, and time period {{period_column}}. Product or segment to analyze: {{product}}. Estimate elasticity:...

You are a SQL specialist in {{sql_dialect}}. Table {{table_name}} stores a hierarchy with id {{id_column}} and parent id {{parent_column}}, plus {{label_column}}. Write recursive C...

You are a marketing analytics lead. Touchpoint data: {{table_name}} with user id, channel, timestamp, and a conversion flag. Compare attribution models: 1. Compute credit per chann...

You are a data warehouse engineer in {{sql_dialect}}. Build a calendar dimension from {{start_date}} to {{end_date}}: 1. One row per day with date key, year, quarter, month, week,...

You are a statistician. Sample: values of {{variable}} from {{dataset_description}}; here is a summary or a sample: {{sample_or_summary}}. Fit a distribution: 1. Suggest candidate...

You are a spreadsheet design expert for {{spreadsheet_app}}. My sheet has columns: {{columns}}. Readers need to spot: {{what_to_highlight}}. Design conditional formatting: 1. Defin...

You are a quantitative analyst. Goal: forecast {{outcome}} for {{horizon}}. Inputs and their uncertainty: {{input_assumptions}} (each with a plausible range or distribution). Build...

You are a data privacy analyst. Here are the column names and sample values of {{table_name}}: {{columns_and_samples}} Produce a PII inventory: 1. Classify each column as direct id...

You are a product analytics engineer in {{sql_dialect}}. Events table: {{events_table}} with user id and event date {{date_column}}. Define day 0 as each user's first event. Comput...

You are a data visualization coach. What I want the reader to understand: "{{message}}". The data: {{data_shape}} (number of categories, time component, parts-of-whole, distributio...

You are a diagnostic analyst. Metric {{metric}} changed from {{before_value}} to {{after_value}} between {{before_period}} and {{after_period}}. Dimensions to slice by: {{dimension...

You are a SQL engineer in {{sql_dialect}}. I want to collapse child rows from {{child_table}} into one row per {{parent_key}}. Provide queries that: 1. Aggregate {{value_column}} i...

You are a vigilant statistician. Aggregate finding: "{{aggregate_finding}}". Possible grouping variable that could reverse it: {{grouping_variable}}. Data context: {{context}}. Che...

You are a forecasting analyst. Series: {{series_description}} at {{frequency}} frequency. Forecast method in question: {{forecast_method}}. Set up a backtest: 1. Recommend a time-a...

You are a finance-savvy analyst. Inputs available: {{available_metrics}} (e.g. ARPU, gross margin, churn, marketing spend, new customers). Build a unit economics model: 1. Derive L...

You are a data ingestion specialist. Here is the head of a messy file from {{source}}: {{csv_head}} Symptoms I have seen: {{symptoms}} (e.g. wrong delimiter, mixed encodings, dates...

You are an EDA analyst. Numeric columns: {{numeric_columns}} from {{dataset_description}}. Build and read a correlation heatmap: 1. Give {{tool}} code for a correlation matrix and...

You are an analytics engineer in {{sql_dialect}}. Events table: {{events_table}} with user id, event timestamp {{timestamp_column}}, and event type. Session timeout: {{timeout}} of...

You are a SQL analyst in {{sql_dialect}}. Table: {{table_name}} with value column {{value_column}} and optional weight {{weight_column}} and group {{group_column}}. Provide queries...

You are a metrics architect. North-star metric: {{north_star}}. Business model in one line: {{business_model}}. Teams that own levers: {{teams}}. Build a metric tree: 1. Decompose...

You are a pragmatic ML lead who pushes back on hype. The business ask: "{{business_ask}}". Data we have: {{available_data}}. The decision the output will inform: {{decision}}. Fram...

You are a data analyst. I have a dataset about: {{dataset}}. Known issues: {{issues}}. Tool: {{tool}}. Give me a cleaning plan: 1. Checks to run first to understand the data (shape...

You are a SQL analyst. Database tables and key columns: {{schema}} Business question: {{question}}. SQL dialect: {{dialect}}. Do this: 1. Restate the question precisely, noting any...

Act as a data scientist doing exploratory analysis. Dataset description: {{dataset}}. Goal of the analysis: {{goal}}. Lay out an EDA plan: 1. The first summary stats and distributi...

You are a data visualization expert. What I want to show: {{message}}. The data: {{data_shape}}. Audience: {{audience}}. Recommend: 1. The best chart type for this message and why....

Act as a data wrangler. Here is a sample of my messy data (first rows and column headers): {{sample}} What I want to do with it afterward: {{goal}}. Produce a cleaning plan: 1. Iss...