THE RESULT
What the data shows
The largest computed metric is 17.1 for AMAZONFEE; the smallest is 0 for B. Metric: credit_to_gross (Ratio).
Ratios compare period credits to period gross value; credits may relate to earlier sales.
| stockcode | gross gbp | credit gbp | lines | credit to gross |
|---|---|---|---|---|
| AMAZONFEE | 13,761.09 | 235,281.59 | 34 | 17.1 |
| M | 78,110.27 | 146,784.46 | 571 | 1.879 |
| 23,843 | 168,469.60 | 168,469.60 | 2 | 1 |
| 23,166 | 81,700.92 | 77,479.64 | 260 | 0.9483 |
| 21,108 | 18,069.28 | 6,591.42 | 270 | 0.3648 |
| 47566B | 11,290.12 | 3,692.95 | 379 | 0.3271 |
| 22,273 | 12,016.34 | 3,512.65 | 344 | 0.2923 |
| 22,191 | 11,971.15 | 2,653.70 | 441 | 0.2217 |
| 20,971 | 13,838.10 | 2,778.77 | 538 | 0.2008 |
| 48,185 | 23,888.81 | 4,554.90 | 372 | 0.1907 |
| 71,477 | 20,658.56 | 3,443.62 | 356 | 0.1667 |
| POST | 78,101.88 | 11,871.24 | 1,256 | 0.152 |
THE METHOD
From source to answer
SQL conditional aggregation with a minimum sales exposure.
The Python source exposes this study’s transformations. The complete project download includes shared preparation and evaluation routines.
THE NEXT DECISION
What follows from the finding
Investigate the product and invoice context before attributing credits to quality problems.
Where the conclusion stops
Historical invoice lines; credits are not reliably matched to original sales. Gross purchases are not profit. Unidentified customers cannot support customer-level conclusions. Exact repeated lines remain unless the study explicitly compares removal.
Related studies may reuse observations or holdouts. These are historical analyses; associations and backtests do not demonstrate commercial impact. Further model tuning needs new, untouched evaluation data.