Seasonality

Sales are down this month. Is something wrong with the business? Or did the month just have too many Mondays?

Examples:
Restaurant with weekend surge | Closed on weekends | Weekends only | Flat

What's happening?

Each month has a different number of Mondays, Tuesdays, Wednesdays, etc. depending on how the week boundaries happen to line up with month boundaries. This alignment changes each year.

If some days of the week tend to have higher sales than others, then these differences affect the total monthly sales amount — even if there is no variability week to week!

In the example above, a constant rate of growth (2% per year) is easily seen in weekly summaries but masked in monthly summaries. The monthly noise may cause decision makers to erroneously conclude that they were responsible for the positive or negative growth.

Analyze monthly summaries cautiously if the underlying data might show weekly seasonality!

Notes

The simulation creates four years of synthetic daily sales data by starting with the first week of specified daily sales and then applying the constant growth rate over time. It then finds the week and month of each data point, computes weekly and monthly totals, and calculates year-over-year growth for the three years that have a year-earlier comparison value.
When there is very low variability (such as in the "Flat" example), you can also see the effect of leap years in the monthly summaries.

the surprising effects of

Weekly Seasonality