# Priya's Q3 Sales — Chapter 1 Worked Example

> Synthetic apparel-retailer Q3 sales scenario used to build Priya's dashboard in Chapter 1 of *The Defensible Decision*. Aggregate rollups only; the figures in the book only visualise aggregates and this dataset is scoped to match.

**Wholly synthetic.** Any resemblance to a real retailer's data is coincidental.

## Files

- [`priya-q3-sales.csv`](priya-q3-sales.csv) — flat aggregate rollups (~15 rows)
- [`priya-q3-dashboard.json`](priya-q3-dashboard.json) — canonical scenario contract with source hash

## What it represents

A single quarter (Q3 2026, July–September) of net revenue at a fictional B2B apparel retailer, broken out three ways:

- **By month** — three monthly totals that sum to the Q3 quarter.
- **By region** — four regions (Central, East, South, West) with an accompanying plan number for each. The plan totals also sum to the Q3 quarter, distributed differently.
- **By product** — six named product lines with revenue, units, cost of goods sold, gross-margin dollars, and gross-margin percentage.

Plus a single prior-quarter (Q2) comparison total.

## Invariants (locked)

The scenario carries executable invariants that the book's release contract enforces. Any change to the numbers must keep every invariant true:

- Q3 net revenue: `$42.70M` exactly.
- Monthly Q3 totals sum to `$42.70M` within one cent.
- Regional Q3 totals sum to `$42.70M` within one cent.
- Regional plan totals sum to `$42.70M` within one cent.
- Product revenue totals sum to `$42.70M` within one cent.
- Q2 baseline: `$41.00M` → Q3 vs Q2 rounds to `+4.1%`.
- Monthly Jul→Sep change rounds to `+4.2%`.
- South is the only region materially below plan (roughly −18.5%).
- South is *not* the smallest region by raw revenue (pedagogically required: alphabetical vs value-ascending sorts both hide South's underperformance without a plan-variance measure).
- Ordering products by gross-margin dollars changes the top product versus ordering by revenue.

## Why the numbers matter

The dataset backs a specific teaching arc: the same underlying quarter, visualised two ways. Chapter 1 shows Copilot's default rendering — a plausible-looking dashboard that quietly deceives on all four panels. The corrected rendering shows what a working analyst produces from the same data once the Part I principles are internalised.

If you tighten the numbers, keep the invariants above intact or the chapter will contradict its own figure.

## What it does NOT represent

- Daily or weekly grain (aggregate rollups only for the current teaching scope).
- Product × region × month grid (the print figure and companion HTML only visualise aggregates; a full grid can be generated on request).
- Customer or SKU data.

## Reproducibility

The dataset ships from the book source at `AlexBooks/applied-ai/ai-directed-viz/data/priya-q3-dashboard.json` and is re-emitted here by `AlexBooks/applied-ai/ai-directed-viz/scripts/generate-priya-dashboard.mjs --site-root <site-root>`. The generator writes the source SHA-256 into `priya-q3-dashboard.json` so drift between the book and this site is detectable.

## License

Released for teaching use alongside *The Defensible Decision*. Use freely.