Seasonality is a recurring pattern tied to a calendar or cycle; a trend is a longer movement. Inspect several comparable cycles, trading exposure and unusual events before describing a trend. A month-to-month change alone cannot distinguish recurring demand from an underlying shift.
Establish which calendar the process follows
The relevant calendar may involve weekdays, school terms, renewal cycles, local holidays or weather seasons. Begin with the process rather than assuming every business follows the same annual pattern. Record dates in the operating time zone and identify partial periods. A month with fewer trading days can have fewer transactions even when demand per open day is unchanged. Total demand and daily workload are different decisions.
Collect enough history to observe repeated cycles if it exists. One rise followed by one fall is insufficient to establish a reliable seasonal pattern. A new product may have no meaningful history, while an acquired business may have incompatible definitions. Mark these limits visibly. Also inspect structural changes such as opening hours, service coverage and measurement rules before treating older periods as comparable evidence.
Compare like periods without concealing important differences
Show the raw series first so readers can see the observed totals. Add a comparison using equivalent trading days or the same holiday-relative window when that better matches the question. Explain the alignment: the first four complete trading weeks after a festival differ from a calendar-month comparison. Moving events can cross month boundaries, so matching month names may still compare different demand conditions.
Do not call every exposure adjustment seasonal adjustment. Dividing by open days answers a narrow rate question. Formal seasonal adjustment estimates recurring components using a model and needs diagnostics, revision policies and sufficient history. A moving average can smooth a chart but may also delay detection of a turning point. State what each transformation does and retain access to the original observations.
Distinguish a recurring pattern from a new operating condition
Review the current result against several previous comparable periods and against the assumptions behind the plan. An unusual promotion, supply interruption or change in customer mix may dominate the seasonal pattern. Ask whether the apparent movement occurs across locations or only where operating conditions changed. Treat plausible explanations as hypotheses until the underlying events and source records support them.
Choose the reporting claim carefully. Demand was lower than the prior calendar month is a direct observation. Demand per open day was unchanged is a narrower comparison requiring a reliable opening calendar. Demand has entered a sustained decline needs more evidence. Record the next comparable observation and the threshold that would prompt action, while considering whether waiting could create an unacceptable service or inventory consequence. A decision note should also say whether the team is responding to observed workload, anticipated demand or an unresolved measurement concern.
A fictional holiday-sensitive repair service
A fictional scenario to explain the method, not a customer result or a market benchmark.
A fictional repair network completes 2,200 jobs during 22 open days in September and 2,000 during 20 open days in October. The monthly total falls by 200 jobs, approximately 9.1%, while both months average 100 jobs per open day. Those figures support different statements about total completed work and daily throughput.
The analyst checks three earlier years and finds that an autumn closure occurs in each, but its dates shift. Rather than infer a downward trend from the two totals, the team compares equivalent open-day windows and checks the backlog. The current October also ends with 120 waiting jobs versus 80 in September, so stable completed work per day does not prove stable demand. Arrivals and available technician hours are requested before changing staffing. All numbers are illustrative, and the opening-day calculation is described as an exposure adjustment rather than a model-based seasonal estimate. The backlog check prevents a stable completion rate from being mistaken for proof that every requested repair was served promptly.
TAKE THIS INTO YOUR NEXT CONVERSATION
A practical next-step checklist.
- Identify the calendar relevant to the service.
- Mark partial periods and exceptional closures.
- Show raw totals beside any exposure adjustment.
- Document history limits before making a trend claim.
Common questions.
Is comparison with last year always sufficient?
No. Holidays, operating coverage and definitions may differ, and one prior year may be unusual.
Does a smooth moving average remove seasonality?
It may reduce visible fluctuations, but it does not by itself identify or validate a seasonal component.
Further reading & source context.
These primary references provide background. They are not evidence for the fictional example above and do not endorse Accuralytix.
Found an error or a missing qualification? Send an editorial correction.
Check my data readiness.
Use a practical tool to turn this idea into a clearer starting point.
Check my data readiness