Analysis that states its assumptions.
Type your question. The Data Analyst preset shapes a prompt that produces a careful analysis, with caveats made explicit.
Ask an AI to "analyse this data" and it confidently invents trends, ignores sample size, and skips the caveats. A clean-looking chart built on a flawed assumption sends the whole decision the wrong way.
The Data Analyst preset pre-tunes the pipeline to produce prompts that demand methodological care: define the question, check the data shape, state assumptions, separate correlation from causation, and quantify uncertainty. Every prompt asks for the analysis approach before the answer.
Type a sentence. Get an expert prompt.
Analyse why our monthly revenue dropped 12% last quarter.
Role: You are a senior data analyst investigating a 12% decline in monthly revenue over the last quarter. Context: The drop could stem from churn, pricing, seasonality, a pipeline issue, or a data artefact. The goal is a defensible explanation, not a confident guess. Treat every claim as a hypothesis to be tested against the numbers. Task: Lay out a structured analysis. Decompose revenue into its drivers (volume, price, mix, retention), propose hypotheses, and specify the cut of data that would confirm or reject each one. Constraints: - State your assumptions explicitly and flag where data is missing. - Distinguish correlation from causation; do not assert a cause without supporting evidence. - Note seasonality and any one-off events that could distort the comparison. - Quantify uncertainty where you can. Format: Return four sections — Revenue decomposition, Ranked hypotheses, Validation plan per hypothesis, and Assumptions and caveats. End with the single analysis you would run first and why.
Tuned for data analyst work.
Explicit assumptions
Prompts force the AI to state what it assumes and where data is missing before drawing conclusions.
Cause vs correlation
Generated prompts insist on separating what moves together from what actually drives the result.
Hypothesis ranking
Each prompt structures the work around testable hypotheses ordered by likelihood.
Uncertainty quantified
Prompts ask for ranges and confidence rather than false precision.
Validation plan
Every prompt closes with the specific data cut needed to confirm or reject each finding.
Three steps, about 100 seconds.
Type one sentence
Ask your question about the data in plain language.
20 agents refine
A pipeline of 20 on-device specialists shapes a rigorous analysis prompt in about 100 seconds.
Copy into your AI
Paste the finished prompt into ChatGPT, Claude, Gemini, or Copilot alongside your data.
Data Analyst preset questions.
Does it analyse the data itself?+
No. It generates the analysis prompt. You paste it, with your data or summary, into your AI assistant.
Can it write SQL or Python for me?+
Yes. Ask for query or code generation in your sentence and the pipeline shapes a prompt that produces it.
Is my data question kept private?+
Yes. Everything runs on-device with Apple Intelligence, so your question and the generated prompt never leave your iPhone.
Does it help with statistical rigour?+
Yes. Prompts built by this preset stress assumptions, uncertainty, and the correlation-causation distinction.
How many prompts can I generate for free?+
Five runs a day on the free tier. Pro removes the limit at £4.99 a month, £29.99 a year, or £59.99 for lifetime access.
Explore related roles.
Build your data analyst prompts on-device.
Free 5 runs a day. Pro from £4.99/mo, £29.99/yr, or £59.99 once. Nothing leaves your iPhone.