Data Analyst — WHAT, WHY, SO WHAT
Data Analyst — decisions, not dashboards
Paste your data; get an analyst-grade HTML report that tells you what's going on and what to do — not just a pile of charts.
🎯 When to use it
You have data (sales, usage, survey, ops) and need someone to read it and tell you the story and the next move — fast, without a BI tool.
⚙️ How it works — the three-layer protocol
1️⃣ WHAT the data shows — the key figures from YOUR numbers: top-line, distribution, highs/lows, changes over time, segments. Quoted, not rounded into mush.
2️⃣ WHY it's likely happening — plausible drivers as labeled hypotheses; correlation vs. causation and confounders called out.
3️⃣ SO WHAT to do — 3–5 concrete, prioritized actions tied to the findings, plus what to measure next.
📊 How it reasons
It infers your columns and grain, computes simple checkable summaries (totals, averages, %s, deltas) from the given numbers, flags data-quality issues, and assembles a self-contained HTML report.
📦 What you get back
A self-contained HTML report (Summary → WHAT → WHY → SO WHAT → results table → Limitations) that opens in any browser, plus [ADD: …] notes where more data would help.
💡 What makes it different
- Insight, not a chart dump — the three-layer protocol gets you to a decision.
- Honest — facts (WHAT) are separated from hypotheses (WHY); it never presents a guess as a measured result, and never invents a number or benchmark.
- Uses only your data — every figure traces back to what you pasted.
👤 Who it's for
Founders, operators, analysts, and PMs who need decisions from their data, fast.
