The October reconciliation is complete: combined contribution after advertising is 2,070.00 CAD on 9 paid orders, but the Course and combined figures are provisional. Refund RF05 (refunds.csv R06) settled in October against a paid Course order with a blank amount, so it is left out of the refund total as policy.md requires. | Measure (CAD) | Workshop | Course (provisional) | Combined (provisional) | |---|---:|---:|---:| | Paid orders | 5 | 4 | 9 | | Gross sales | 1,920.00 | 900.00 | 2,820.00 | | Applied refunds | 240.00 | 60.00 | 300.00 | | Net revenue | 1,680.00 | 840.00 | 2,520.00 | | Advertising spend | 300.00 | 150.00 | 450.00 | | Contribution after advertising | 1,380.00 | 690.00 | 2,070.00 | | Net revenue per paid order | 336.00 | 210.00 | 280.00 | - **Range on the provisional figures:** RF05 can be at most the 360.00 order value. At that ceiling, Course net revenue would be 480.00 and combined contribution 1,710.00. This is a bound, not an estimate. - **Orphan refund:** RF04 (R05, 120.00) is settled against order N9999, which is not in sales.csv. It is flagged and excluded; if it belongs to an October paid order, net revenue falls by 120.00. - **Workshop refunds outside the period:** RF03 (240.00, pending) and RF06 (240.00, settled 2 November) relate to October Workshop orders. Neither reduces October under the policy. - **Recommendation:** do not shift ad budget between products on this month's data. Both return 5.60 of net revenue per ad dollar, and the Course figure can only fall once RF05 is known. Resolve RF05 and RF04 before treating October as final. - **Not profit:** the contribution figure is net revenue less ad spend only. No COGS, payment fees, salaries or overhead were supplied. **Reconciliation** | Source | Rows | Included | Excluded | Included amount | Excluded amount | |---|---:|---:|---:|---:|---:| | sales.csv | 15 | 9 | 6 | 2,820.00 | 1,320.00 | | refunds.csv | 7 | 2 | 5 | 300.00 | 660.00, plus R06 unknown | | ads.csv | 7 | 5 | 2 | 450.00 | 180.00 | All 13 excluded rows are in `exceptions.csv` with the reason and policy basis for each. **Assumptions** - Cleaned files hold only the records counted in the totals; every other source row is in `exceptions.csv`. R06 is there with its amount left blank rather than in `cleaned-refunds.csv`. - An order's month comes from its `date` column, not `updated_at`. - A refund's product is taken from its order, since refunds.csv has no product column. **Checks performed** - `reconcile.py` ran cleanly, and its built-in checks passed: every source row lands in exactly one of cleaned or exceptions, included plus excluded amounts equal the source totals for each file, and Workshop plus Course equals combined. - A separate script recomputed all headline figures from the cleaned CSVs alone and matched. - That script also confirmed IDs are unique, all included records are paid or settled and dated in October, and no applied refund exceeds its order's gross. - I read back `summary.md` before two small wording edits and all four CSVs. After the edits I confirmed the summary's key figures and the changed wording by script, not by a full re-read. **Limitations** - Nothing was verified against a bank, payment processor or ad platform; results rest only on the three CSVs and policy.md. - sales.csv has no channel field, so revenue cannot be tied to search or social. - The inputs give no refund reasons, so the cause of the Workshop refunds is unknown. **Estimated human time to finish:** about 15–20 minutes to review the summary and exceptions, plus however long it takes to obtain the RF05 amount and trace N9999. This is an estimate, not a measurement. Files are in `output/`: - summary.md - cleaned-sales.csv - cleaned-refunds.csv - cleaned-ads.csv - exceptions.csv - reconcile.py