A working spreadsheet for GCC planners: re-index your history onto the clock the demand actually uses, derive your own phase factors, and work back from Ramadan 1448 to the date your purchase orders have to leave.
Free. You also get The Corridor — one corridor decision a week. Unsubscribe any time.
The Islamic calendar is lunar. Twelve lunar months run about 354 days against the Gregorian 365, so every Islamic date drifts roughly eleven days earlier each year. Ramadan 2025 began in early March. In 2026, mid-February. In 2027, around 8 February.
Holt-Winters, SARIMA and the seasonality components inside most ERP demand modules learn a repeating pattern indexed to a fixed calendar position — month 3 behaves like month 3 did last year. Feed them four years of GCC retail history and they will dutifully learn that March is a high month, because Ramadan happened to fall there once.
This is not a tuning problem. The parameters are attached to the wrong axis. No amount of parameter tuning fixes a model whose clock disagrees with the demand’s clock.
And the second mistake: treating Ramadan as one lift factor — “food goes up 30%” — applied to the whole month. The real signature has four phases with different category behaviour, and the largest volume movement usually happens before Ramadan begins.
Paste two years of weekly demand, enter each year’s Ramadan start, and every week is re-indexed onto days-from-Ramadan with automatic phase labelling. If your peaks only line up on this axis, your seasonality index is being fitted to noise.
Derive your own factors per category across all four phases. A borrowed uplift factor is a guess wearing a decimal point.
Twelve milestones driven off one editable date cell — from PO placement back-calculated on an Asia–Jebel Ali lane, through on-shelf, to the post-Eid trough.
High-value, low-velocity SKUs whose demand sits in two or three events a year get ranked C, then have their peaks erased by a flat index. One test catches them before the rules do.
Working back from Ramadan 1448 on an Asia–Jebel Ali lane. Substitute your own measured transit — these are illustrative.
| Milestone | Date |
|---|---|
| PO placement — the real deadline | 19 Oct 2026 |
| Forecast locked, S&OP signed off | 16 Nov 2026 |
| DC receipt — available to promise | 25 Jan 2027 |
| On shelf, displays built | 27 Jan 2027 |
| Ramadan begins | 8 Feb 2027 |
| Eid al-Fitr | 9 Mar 2027 |
Every date is ±2 days pending moon sighting. The PO row is the one that cannot slip. If you are reading this after it, you are managing a shortfall rather than planning a peak — the Uplift Worksheet will still size the gap.
Free, no course, no upsell. You also get The Corridor — one corridor decision a week.
Read the full reasoning first: Why your seasonality model breaks every year · The C-class trap