THE RESULT
What the data shows
The reconciled model preserves 541,909 source lines and £10.67 million of gross sales. Credits are 8.41% of gross invoiced value; identified customers cover 83.5% of gross sales.
The model makes grain, signed credits, distinct invoices, customer coverage, and filter context explicit. This is a BI implementation over existing source data, not an independent replication.
| Month | CompleteMonth | GrossSales | Credits | NetValue | SourceLines | CreditValueRatio |
|---|---|---|---|---|---|---|
| 2010-12 | Complete month | 823,746.14 | 74,789.12 | 748,957.02 | 42,481 | 0.09079 |
| 2011-01 | Complete month | 691,364.56 | 131,364.30 | 560,000.26 | 35,147 | 0.19 |
| 2011-02 | Complete month | 523,631.89 | 25,569.24 | 498,062.65 | 27,707 | 0.04883 |
| 2011-03 | Complete month | 717,639.36 | 34,372.28 | 683,267.08 | 36,748 | 0.0479 |
| 2011-04 | Complete month | 537,808.62 | 44,601.50 | 493,207.12 | 29,916 | 0.08293 |
| 2011-05 | Complete month | 770,536.02 | 47,202.51 | 723,333.51 | 37,030 | 0.06126 |
| 2011-06 | Complete month | 761,739.90 | 70,616.78 | 691,123.12 | 36,874 | 0.0927 |
| 2011-07 | Complete month | 719,221.19 | 37,921.08 | 681,300.11 | 39,518 | 0.05273 |
| 2011-08 | Complete month | 759,138.38 | 54,333.75 | 704,804.63 | 35,284 | 0.07157 |
| 2011-09 | Complete month | 1,058,590.17 | 38,902.55 | 1,019,687.62 | 50,226 | 0.03675 |
| 2011-10 | Complete month | 1,154,979.30 | 84,274.63 | 1,070,704.67 | 60,742 | 0.07297 |
| 2011-11 | Complete month | 1,509,496.33 | 47,740.08 | 1,461,756.25 | 84,711 | 0.03163 |
Inside the Power BI project
Three editable report pages, 17 DAX measures, and a documented model with active single-direction relationships.
- Retail performance
- Customer lens
- Data quality & reconciliation
The native report format passes Microsoft’s validator. Data checks and visual field references pass independently. Desktop refresh and rendering remain to be verified; the chart above is a Python analysis figure.
Download all DAX measures ↓Download the Power BI runbook ↓
THE METHOD
From source to answer
Invoice-line fact table, four conformed dimensions, single-direction relationships, 17 filter-aware DAX measures, and independently reconciled source totals.
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
Use the report to select a country and period, inspect credit exposure and customer coverage, then define a measurable retention experiment with a commercial owner.
Where the conclusion stops
This report extends the original retail case study using the same historical observations. Snapshot customer segments are fixed at 10 December 2011. Credits are not linked to original sales, and invoice value is not profit. Native Power BI refresh and rendering remain unverified because Desktop is not installed.
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.