The task and original case
Build a small demand forecast that treats a promotion week as a different scenario. Your portfolio should show what the average includes before presenting a target. The invented history supports a calculation, not a proven forecast for a real store.
| Case element | Value or requirement |
|---|---|
| Inputs | Ordinary weekly demand 10,12,8; a promotion week is 30. |
| Build | A scenario forecast with an explicit promotion assumption and error-review loop. |
This is an original invented exercise. It is independent practice, with no employer assessment or open-vacancy claim. Download the inputs, attempt your own result, then open the separate answer key.
Data and worked inputs
Build a result another person can follow
Label three ordinary weeks with demand 10,12 and 8, and one promotion week with demand 30. Store the week type as a field instead of excluding a row without explanation. Calculate an ordinary-week baseline from the three ordinary observations, and an all-weeks summary from all four.
State which upcoming scenario you are planning: an ordinary week or a promotion-like week. If you use a PivotTable, inspect the underlying sums and counts and select the intended summary function; Microsoft's documentation covers the field setting. Your planning choice must come from a declared scenario assumption, not from the software's default aggregation.
Sources[1]
Check the worked result
Ordinary demand totals 30 across three weeks, giving a mean of 10. All four weeks total 60, giving a mean of 15. The extra five units in the overall average reflect inclusion of the promotion observation; they are not evidence that ordinary demand grew by 50%.
For an ordinary-week practice plan,10 is the stated historical baseline. A future promotion scenario needs its own assumption. One observed promotion week does not establish a stable promotion effect, seasonality or a reliable probability of demand 30.
Make the failure visible
Remove the promotion label while retaining 30 and see whether the report still explains the mixed history. Test a missing demand value and avoid dividing by the full number of weeks when one observation is unknown. Also distinguish units requested from units sold if a later exercise introduces stockouts; fulfilled sales can understate demand.
Revise it and ask someone to reproduce it
Add a new ordinary week with demand 14. The ordinary baseline becomes 44 ÷4 =11. Compare that update with the previous baseline and record the error of an earlier 10-unit prediction if you made one. Ask a peer to choose which historical rows belong in a future ordinary-week scenario and defend their choice.
- Week types and included observations are visible.
- Ordinary and mixed-history averages reconcile to their sums and counts.
- The recommendation states a scenario and avoids claiming a proven promotion effect.
Describe the work accurately
After completing your own version, you could adapt this project bullet: “Built an independent demand scenario worksheet separating ordinary and promotion observations and documenting forecast assumptions and later errors.”
Use Projects or Independent practice as the heading. Keep the input files, result and revision together. Describe only your own completed work; the invented numbers are model outputs, not employer results or measured business impact.