Planning around Diwali: the event uplift your spreadsheet can't learn
Festival demand is not noise to be smoothed away — it is signal to be graded. Why a scheme calendar beats a seasonal index for Indian demand.
Ask a standard forecasting model about Diwali and it does one of two things: it smooths the spike into the baseline (so you under-stock every year), or it bakes last year's exact spike into a seasonal index (so you over-stock when the festival shifts by three weeks on the calendar, which it does).
Neither is planning. Both are curve-fitting pretending to understand your business.
Events are objects, not bumps
In SangamIQ, a Diwali scheme is a first-class object on a calendar: a window, an expected uplift, and — crucially — a grade after the fact. When the window closes, we compute the realized uplift against the same window last year and score how well the plan called it. That grade feeds the next cycle.
| Seasonal index | Event calendar | |
|---|---|---|
| Handles shifting dates | ✗ | ✓ |
| Separates scheme from baseline | ✗ | ✓ |
| Learns from what actually happened | ✗ | ✓ (realized-uplift grade) |
Why this matters for Indian ops
Indian demand is scheme-driven in a way that annual seasonality can't capture: distributor schemes, GST-timing effects, regional festival calendars, monsoon-linked categories. A model that treats all of that as one smooth sine wave will be wrong in a specific, expensive direction — cash locked in the wrong stock while the fast movers stock out.
Grade the guess, and next Diwali you're arguing about a number with a track record instead of a hunch.
Every number in SangamIQ shows its work — see our live accuracy page, or start free and put your own CSV through it in ten minutes.