Seasonality, Events and the Calendar
Most of what looks like unpredictable demand is on a calendar somebody could have built. Building it once removes a year of surprises.
Operations rediscover the same events every year. A maintained calendar converts recurring surprises into planned weeks.
The practical lesson in “Seasonality, Events and the Calendar” is to connect every record to a named decision. Organisations exploring Monitask for productivity software for business can add structured workforce context, provided the use is disclosed and the interpretation is reviewed with the people affected.
What belongs on it
School terms and holidays, which drive demand in most consumer operations.
For an independent reference related to “Seasonality, Events and the Calendar”, consult the OSHA worker-fatigue resources; it provides a useful external check on scheduling, working-time and workforce-planning assumptions.
Public holidays, including the weeks around them.
Paydays and benefit payment dates, which move weekly demand in many settings.
Local events: matches, festivals, conferences, markets.
Your own calendar: promotions, price changes, openings, closures, system work.
Building it once
Last year's demand by day, with the top twenty and bottom twenty days marked.
Then work out what each was.
Most of them have an identifiable cause and most of those recur, which is the whole exercise.
An afternoon, once, and it carries forward every year with small updates.
The pattern around holidays
The days before and after frequently matter more than the day itself.
Christmas eve, the day before a long weekend, the first day back.
These are predictable and are the ones most often staffed as ordinary days.
The internal calendar problem
Marketing runs a promotion and the schedule was built a fortnight earlier.
Which is an information flow problem rather than a forecasting one.
One standing request — tell the scheduler about anything that changes demand, four weeks out — fixes a surprising share of forecast misses, and it costs nothing.
Seasonality proper
Some operations have a strong annual shape: retail, hospitality, agriculture, education, tourism.
Others have almost none.
Know which you are, from your own data, rather than assuming — and if you have a shape, the staffing model should have one too.
The multi-year view
Three years of data shows which patterns are real and which were one-offs.
Two years is enough to start.
One year tells you very little, and most operations are working from one year or from memory.
What to do with it
Build the schedule from the base shape plus the calendar.
Flag the weeks that need attention, four weeks out.
And review the calendar annually, adding what you learned, which is the two-line record the previous note recommends.
What to check
Do you have an events calendar, written down?
Does it include your own promotions and changes?
Who tells the scheduler about a demand-affecting decision, and how far ahead?
And did last year's top twenty days have identifiable causes?