Best data analysis AI Prompts

Prompt - Data Storytelling for a Presentation

Act as a data storyteller. I need to present these findings: [PASTE KEY DATA/INSIGHTS] to [AUDIENCE]. Help me build the narrative: the one core message they should remember, the 3 supporting points in a logical order, which chart type fits each point, the 'so what' business implication, and how to open and close for impact. Turn numbers into a story, not a data dump. Suggest where a single striking stat should lead.

Prompt - KPI and Metrics Framework Designer

Claude

Prompt - KPI and Metrics Framework Designer

Act as an analytics advisor. Help me define the right metrics for [BUSINESS/TEAM/GOAL]. Distinguish between the one North Star metric, the input metrics that drive it, and vanity metrics to ignore. For each key metric: define it precisely, why it matters, how to measure it, and a healthy target range to aim for. Warn me about metrics that can be gamed. Keep the set small enough to actually act on.

Prompt - Python Data Script Generator

Prompt - Python Data Script Generator

Act as a Python data engineer. Write a clean, commented Python script that [TASK, e.g. reads a CSV, cleans it, and outputs a summary]. Input format: [DESCRIBE]. Desired output: [DESCRIBE]. Use pandas where sensible, handle missing values and errors gracefully, and print a short summary at the end. Explain how to run it and what libraries to install. Keep it readable for someone who's intermediate at Python.

Prompt - Cohort and Retention Analysis Explainer

Claude

Prompt - Cohort and Retention Analysis Explainer

Act as a product data analyst. Here is my user or revenue data by cohort: [PASTE OR DESCRIBE]. Analyze retention: identify which cohorts retain best and worst, where the biggest drop-off happens, likely reasons behind the pattern, and 3 hypotheses to improve retention that I could test. Explain any metric you use in plain language and flag where the data is too thin to conclude anything.

Prompt - Excel/Sheets Formula Builder

Prompt - Excel/Sheets Formula Builder

Act as a spreadsheet expert. I need a formula in [Excel/Google Sheets] that [DESCRIBE WHAT YOU WANT]. My data is laid out as: [DESCRIBE COLUMNS/RANGE]. Give me the exact formula, explain what each part does, show a worked example, and offer one alternative approach (e.g. a pivot table or a cleaner formula) if it's better. Warn me about common errors like circular references or off-by-one ranges.

Prompt - Dashboard Narrative and Report

Claude

Prompt - Dashboard Narrative and Report

You are a data storyteller. Here are my key metrics for [PERIOD]: [PASTE METRICS]. Write an executive-ready narrative: lead with the single most important takeaway, explain what changed and why it likely happened, highlight one risk and one opportunity, and end with 3 recommended actions. Keep it under 300 words, plain language, no jargon, and quantify claims where the data supports it.

Prompt - Clean and Structure Messy Data

Prompt - Clean and Structure Messy Data

Act as a data analyst. Here is messy data: [PASTE DATA]. Clean and structure it: identify inconsistencies (formats, duplicates, missing values, outliers), propose a cleaning plan, and return a tidy version. Explain each transformation you'd apply and flag any values you're unsure about rather than silently changing them. Suggest the ideal column schema for analysis.

Prompt - Data Insight and Action Extractor