# Example: pasted CSV analysis

## Filled inputs

- `{{question}}`: Which product category drove the drop in revenue in September compared with August?
- `{{csv_data}}` (first rows shown):
  ```
  order_id,order_date,category,units,unit_price,region
  10021,2026-08-01,Accessories,3,19.00,North
  10022,2026-08-01,Footwear,1,89.00,South
  10023,2026-08-02,Apparel,2,45.00,North
  ...
  ```
- `{{data_notes}}`: One row per order line. Prices in USD, excluding tax. Refunds are not included.
- `{{audience}}`: regional sales managers

## Sample output (trimmed)

### Data check
- 1,412 rows, 6 columns. Dates run 2026-08-01 to 2026-09-30.
- No missing values in units or unit_price. 4 rows have region blank (0.3% of rows); kept and shown as "Unknown".
- Revenue is computed as units x unit_price (there is no revenue column).
- Counts computed with the code tool.

### Answer
Footwear accounts for most of the decline. Total revenue fell from $48,210 in August to $41,935 in September (-$6,275). Footwear fell by $5,140, which is 82% of the total drop, mainly in the South region (-$4,320).

### Supporting table
| Category | Aug revenue | Sep revenue | Change |
|---|---|---|---|
| Footwear | ... | ... | -5,140 |

### Caveats
- Refunds are excluded, so net revenue may differ.
- Two months only; this cannot tell you whether September is unusual or seasonal.

### Next questions worth asking
- Did Footwear units fall, or prices (mix or discounting)?