STOP 02 / 31One line. Several stories.
Layout
STOP 02 · Foundations

One line. Several stories.

EXPLORE~6 min
Back to route map
  • Northstar’s sales are growing, fluctuate across weekdays, and jump around Black Friday every year.
  • The observed line mixes all of these with randomness.
  • Change the ingredients below to see how they combine.
Sales components at Northstar. Illustrative Northstar examples, not measured business results.
ComponentWhat it isNorthstar example
Trend
  • More customers and repeat purchases can lift the store’s baseline over time.
  • Trend need not be a straight line.

More repeat buyers could lift Northstar's ordinary-day sales over several years, even after removing the usual holiday shopping ramp.

Seasonality
  • Weekly shopping patterns and the annual holiday season repeat.
  • Black Friday is an annual calendar event, but its exact date moves; model its event window explicitly.

Weekly shopping habits repeat, while Black Friday needs its actual moving date and event-relative window rather than a fixed November date.

Cycle
  • A rise and fall over a non-fixed period, such as an economic cycle.
  • Unlike seasonality, it is not tied to a reliable weekday, month, or annual calendar slot.
  • A prolonged consumer-spending slowdown could reduce Northstar's sales for several quarters.
  • Its duration would not be a fixed calendar season; this is a hypothetical business example, not a separate fitted simulator component.
Remainder
  • What is left after the chosen components are removed: random variation, outages, one-off campaigns, and patterns the decomposition missed.
  • Unexplained does not automatically mean pure white noise.
  • An unexpected Northstar checkout outage can lower sales for a day.
  • A repeatedly missed Black Friday spike, however, suggests an omitted event effect rather than meaningless randomness.

How seasonal effects scale

How seasonal effects scale. Illustrative Northstar examples, not measured business results.
PatternWhat it isNorthstar example
Additive

A seasonal effect adds a fixed amount, regardless of the current business level.

A hypothetical Northstar weekend adds $2,000 whether ordinary sales are $20,000 or $40,000.

Multiplicative
  • A seasonal effect adds a percentage, so dollar swings grow with the business.
  • Logs of positive values can help stabilize proportional variation.

A hypothetical 10% Northstar weekend lift adds $2,000 to a $20,000 base but $4,000 to a $40,000 base.

observed sales = trend + weekly pattern + annual pattern + event + remainder
FIG 02.1

Build sales from the ingredients

WHAT THIS SHOWS

Change the ingredients of Northstar's simulated daily sales and compare the observed line with the supplied underlying components.

The question it answers: Which changes persist, which repeat, and which are random or event-driven?

How effects scale
Interactive time series chart$11k$44k$76k$109kJan ’23Jul ’23Jan ’24Jul ’24Jan ’25Jul ’25Dec ’25
Observed salesUnderlying trend
WEEKLY EFFECT · FIRST 4 WEEKS
Interactive time series chart−$2k−$1k$1k$2k
Weekly component
EVENT EFFECT · THREE YEARS
Interactive time series chart−$7k$16k$39k$61k
Black Friday + Cyber Monday

Set annual growth to zero. The holiday spikes still return. Now turn Black Friday lift off: permanent growth and recurring events are different ingredients. All component values shown here are known from the simulator.

Fine print & assumptions

Constructed teaching data and illustrative calculations, not a production forecasting service. Read the diagram's method notes and distinguish supplied simulator components from fitted estimates.

CHECK YOURSELF · CHOOSE AN ANSWER

Northstar’s Black Friday peak returns every year. What is it?

Your learning note

Private · just for you