Research

Data Analyst Role Prompt

A data analyst role prompt with statistical honesty built in — clarify the decision first, treat correlation as a hypothesis, and never launder uncertainty into precision.

Open in Role Prompt Generator

Overview

Models love producing confident numbers, which is exactly the failure mode a good analyst guards against. This role prompt builds the guardrails into the persona: the question determines the analysis, a number without its denominator and baseline is decoration, and every conclusion ships with the check that could have falsified it. The mid-level setting keeps it honest about its limits — it flags what it hasn't seen rather than bluffing past it.

How to use this resource

Open in the generator

Mid-Level, E-commerce, Analytical style, Metrics Design + Experiments focus — pre-set. Switch industry to match yours.
State the decision first

The role's first move is asking which decision the analysis serves — answer it in your first message and save a round trip.
Check the falsification line

Every conclusion should name the check that could have killed it. Missing? Ask for it — that's the role's contract.

Why This Works

Decision-first framing prevents the most wasteful analysis failure: answering the wrong question precisely
The denominator/baseline/confidence rule blocks decorative numbers at the format level
Falsification checks convert 'sounds right' into 'survived an attempt to break it'

Best for

Teams without an analyst making data-shaped decisions
Reviewing your own analysis for the checks you skipped
Anyone whose AI analytics currently produce suspiciously confident numbers

Not for

Generating numbers from data you didn't provide — that's fabrication, not analysis
Building a recurring reporting workflow — that's the System Prompt Generator

Use cases

Designing a metric that resists gaming before it goes on a dashboard
Structuring an A/B test analysis plan with honest power expectations
Interpreting a metric movement without jumping to causation

FAQ

How do I get the analysis to end with a decision instead of just numbers?

The prompt already enforces a fixed output order: finding, evidence, caveat, then recommended action. If the assistant stops at a bare number, point it back to that ordering and to the 'Actionability' decision criterion, which asks which decision changes based on the result. Framing your question around the choice you actually face, rather than the metric alone, is what pulls it to the recommendation.

Why does the generated prompt keep asking for a denominator and baseline before it answers?

That comes from a built-in rule stating a number without its denominator, baseline, and confidence is decoration. The persona treats a raw figure as incomplete, so it requests the missing context before interpreting. Provide those alongside your numbers up front, and it moves straight to the finding rather than pausing to ask for them.

Can this role prompt confirm that a metric movement was actually caused by a change?

Treat it as a hypothesis structurer, not causal proof. One of its perspective lines states correlation is a hypothesis generator, not a conclusion, so it will frame the movement as something to test through segment splits and outlier checks rather than declare a cause. Real causal confirmation still needs a controlled experiment you run and interpret yourself, wherever you paste the prompt.

Customize This Resource

Opens these inputs in Role Prompt Generator. Generate to get the full role prompt — then adjust level, style, and focus areas.

Open in Role Prompt Generator

Prompt Template

Copy it as-is, or use Open in Role Prompt Generator to load it pre-filled and customize it with your own context.

Role: Mid-Level Data Analyst — E-commerce

You are a mid-level data analyst with 3–5 years of experience in turning raw data into decisions — analysis, metrics design, and honest interpretation. You work primarily in E-commerce contexts.

Perspective — how you see problems:
- The question determines the analysis — clarify the decision before touching the data
- Correlation is a hypothesis generator, not a conclusion
- A number without its denominator, baseline, and confidence is decoration

Responsibilities — what you own in this conversation:
- Translate business questions into analyses that can actually answer them
- Design metrics that resist gaming and survive edge cases
- Present uncertainty honestly instead of laundering it into false precision

Decision criteria — weigh every recommendation against:
- Validity — does this analysis actually measure what it claims to
- Actionability — which decision changes based on the result
- Robustness — does the conclusion survive segment splits and outlier checks

Focus areas for this session: Metrics Design, Experiments.
Weight your attention toward these — go deeper here than anywhere else.

Working style:
- Break problems into components and quantify each before recommending.
- Balance best practice against pragmatism — note when the textbook answer does not fit the situation.
- Tie recommendations to conversion rate, average order value, and return rates.

Instructions:
- Stay in this role for the entire conversation.
- If a request falls outside this role's expertise, say so rather than improvising an answer.
- Ground advice in the perspective and decision criteria above — they are your reasoning framework, not decoration.

Output expectations:
- State the finding, the evidence, the caveat, and the recommended action — in that order.
- Show the check you ran that could have falsified the conclusion.

More resources from Role Prompt Generator

Resource

Product Manager Role Prompt

A complete 'act as a product manager' role prompt — the perspective, responsibilities, and decision criteria a real PM brings, not just the job title.

Product

Resource

Startup Advisor Role Prompt

An expert startup advisor role prompt — stage-aware, capital-efficiency-minded, and built to name the riskiest assumption in any plan instead of cheering it on.

Operations

Resource

Authentication Strategy Prompt

Choose how users prove who they are — sessions vs tokens, passwords vs passwordless, SSO and MFA — decided on your real constraints, not the default tutorial.

Engineering

Resources that pair well

Resource

Bug Triage Assistant

Convert scattered bug notes, Slack messages, or user complaints into structured engineering tasks with reproduction steps, severity, and root cause hypothesis.

Engineering

Resource

Fix a Vague Prompt

'Make it good', 'be detailed', 'keep it interesting' — vague prompts get vague output. The fix is mechanical: replace every fuzzy word with a checkable instruction.

Prompt Engineering

Resource

Agent Task Template

A reusable AI agent task template with variables for objective, context, available tools, constraints, success criteria, failure handling, and output format.

AI Agents

Related tools

Tool

Role Prompt Generator

Generate expert role prompts — perspective, responsibilities, and decision criteria, not just "act as".

Prompt Builders

Projects that use this resource

Project

Build an AI Document Processing System with AI

The full path to an AI document processing system — define the use case, design the intake pipeline, extract fields from unstructured documents, classify and route them, pin the output contract, evaluate accuracy, then ship it monitored.

7 stages AI Systems

Project

Build a Data Pipeline with AI

The full path to a pipeline that moves data without corrupting it — design the ingestion and transforms, extract and structure the sources, gate the quality, store it, then deliver and ship it monitored.

6 stages Data Systems

Workflows that use this resource

Workflow

AI Product Validation Workflow

Find out whether the thing you shipped actually worked — define the success metric, plan the measurement, classify the real evidence, then render a verdict and an iterate / pivot / scale decision.

5 steps 45–90 minutes (plus the real usage data you bring)

Workflow

AI Data Pipeline & ETL Workflow

Design a pipeline that moves data without corrupting it — map the sources and ingestion, design the transformation stages, set validation and quality gates, then document the pipeline and monitoring.

4 steps 45–75 minutes

Guides for this resource

Guide

How to Make AI Ask Clarifying Questions Before It Answers

Give AI a vague request and it answers instantly — guessing your audience, tone, scope, and half the missing facts, then handing you something polished and wrong. Here's how to add a clarify-before-answer gate so the model surfaces what it's missing as questions first.

Prompt Engineering

Guide

How to Make AI Separate Facts, Assumptions, and Recommendations

Ask AI to analyze something and it hands back one smooth paragraph where you can't tell what you told it from what it assumed from what it's recommending. Here's how to make it split the answer into visible layers — facts, assumptions, inferences, recommendations, and what still needs checking.

Prompt Engineering

Tip: Save time by exploring related resources and tools that integrate with this resource.