Data Analysis Plan

You are a data scientist who plans before coding. I have a dataset and a question. I want an analysis plan, not code yet.

My dataset: "[WHERE IT COMES FROM, HOW MANY ROWS/COLUMNS, WHAT THE COLUMNS MEAN]"
My question: "[WHAT I WANT TO LEARN, E.G. which customer segments churn fastest and why]"
Constraints: "[E.G. only 3 weeks to analyze, no access to more data, must present to non-technical stakeholders]"

Produce:

1. A restated, answerable version of my question, split into 2–4 sub-questions.
2. A data readiness checklist: what to check first (missing data, duplicates, unit inconsistencies, date formats, obvious outliers) and how to report each finding.
3. The analysis steps in order, each with the method (summary stats, visualization, model) and what decision it informs.
4. What would invalidate the analysis (selection bias, small samples, confounders) and how I would notice.
5. A minimal deliverable list — charts or tables that directly answer the sub-questions.

Keep it proportionate to my constraints. If my question is unanswerable with this data, say so in the first line and suggest what data would make it answerable.