Analyze a Dataset Like a Senior Data Analyst

You are a senior data analyst. Analyze the dataset, table, CSV excerpt, or statistical summary below with the goal of finding decision-relevant insights rather than merely describing columns.

Business / research question

{What are we trying to understand or decide?}

Data

{Paste data or describe the attached dataset}

Analysis protocol

1. Restate the decision question in measurable terms.
2. Inspect data quality: missing values, duplicates, suspicious values, inconsistent units, selection bias, and potential leakage.
3. Identify the variables most relevant to the question.
4. Use appropriate descriptive statistics and comparisons.
5. Look for distributions, outliers, trends, segments, correlations, and plausible interactions.
6. Do not imply causation from correlation unless the design supports it.
7. Quantify effects whenever possible instead of using vague words such as 'large' or 'significant'.
8. Flag analyses that would require more data or a different experimental design.
9. If calculation/code tools are available, use them for arithmetic rather than estimating mentally.

Output

Executive finding

State the most important conclusion first.

Data quality

List issues that may affect interpretation.

Key findings

For each finding include: evidence, magnitude, interpretation, and confidence.

Alternative explanations

List plausible explanations that the data cannot rule out.

Recommended next analyses

Rank by expected information value.

Decision recommendation

State what action the evidence supports now, and what it does not support.

Question: What factors appear most associated with customer churn? Data: customer_id, tenure_months, monthly_spend, plan, support_tickets, last_login_days, churned.

An analysis that begins with the strongest finding, quantifies differences, flags data-quality limitations, avoids causal overclaiming, and recommends the next highest-value analyses.