Supported finding
Total actual is 374,369 against a 369,025 target, or 101.45% attainment. Yet Monday is 12.31% below its pooled weekday target and Thursday is 10.88% below. Attainment divides sums rather than averaging cell percentages.
07 / Business intelligence
MA-SET1 business intelligence
A workbook rebuilt as a reproducible analytical case. Explore targets, marketing, store sales, customer history, recorded category values and brand ratings.
The pooled target hides weekday shortfalls.
Question
Does overall target attainment describe every weekday in the Sparkline module?
Fixed comparison: Sparkline SALE/TARGET matrix, all 11 cities and 77 city/weekday pairs; period and monetary units unspecified. Controls below do not change this summary.
Decision implication
Before using the overall result, a business analyst should inspect the weekday and city rows for offsetting shortfalls and surpluses. Keep this undated module separate from Store geography and other workbook populations.
Limitation
A single undated matrix with unspecified monetary units. The source does not establish causes, a recurring weekly pattern or a forecast. No cross-module customer or geographic join is assumed.
Actual sales
against target.
Each module uses its own population and units. Customer and geographic links are not assumed across sheets.
Start here Start with all cities, then select one city or another workbook module. Each module has its own scope.
Loading workbook aggregates…
Attainment by weekday
Exact values are in the table below.
What the numbers support
Preserve the grain.
Make the limits visible.
Fresh calculations from the source workbook. The earlier report and deck remain historical artifacts and are not recertified by this edition.
- Source validation
- The exact 1,012,447-byte workbook is checksum-bound. Schema, missing values, ranges, transaction uniqueness, target alignment and every Store revenue identity are checked. Original cells remain unchanged.
- Customer identities
- RFM uses C-prefixed text IDs; Pricing uses numeric IDs. There are zero exact typed-key matches. Formatting numbers into similar-looking codes does not establish identity. No RFM–Pricing join or cross-module correlation is made.
- Time and units
- Website months have no year. Store dates extend through 2083 on an artificial analytical calendar. Sparkline has no dated week. Only Website spend is labelled USD; other monetary-looking fields retain unspecified source units.
- Weighting and recency
- Target attainment divides total actual by total target. RFM recency uses 12 April 2026, one day after the latest source transaction. Frequency counts transactions; monetary value sums recorded amounts. Cohort amounts cover the full source window for customers in each recency group.
- Source rights
- Acquisition date and redistribution rights are unverified. Authored calculations and anonymous aggregates are packaged; the original workbook and customer-level records are excluded. Full replay requires authorized exact-source access.
Recreate all six modules
Verify the package without the workbook. With an authorized matching workbook, regenerate every aggregate and compare it with the saved baseline.
Included
Python code and aggregates included. Full replay needs the original workbook.
Software
Python with the pinned packages; no Excel installation needed for the replay.
Full replay
Authorized exact workbook required; customer-level records are excluded.
Reproduction package
Python pipeline, six aggregate CSVs, browser JSON, dependencies, tests, checksums and source instructions.
Download reproduction ZIPZIP · 19 KBMethods and definitions
Module grains, formulas, source periods, units and boundaries.
Download method notesMD · 4 KBVerify and replay
python -m pip install -r requirements.txt python verify.py python verify.py --input YOUR_WORKBOOK
Run inside the extracted package. Full replay requires exact DATA_MA_SET1.xlsx bytes; a mismatch stops the run.