Retail hiring rises around the holidays. School schedules change employment patterns. Some prices move regularly with seasons. These recurring movements can obscure what changed beyond the normal calendar pattern. Seasonal adjustment is a statistical attempt to make that underlying movement easier to see.
At a glance
- Adjusted series help with short-term comparisons affected by recurring patterns.
- Unadjusted figures retain the observed seasonal movement.
- Do not mix adjusted and unadjusted observations in the same calculation.
What the adjustment does
BLS describes seasonal adjustment as a way to estimate and remove recurring seasonal influences from a time series. The resulting series can make month-to-month movements easier to interpret. It does not turn a sample estimate into perfect knowledge or remove every unexpected event.
The appropriate series depends on your question. A manager planning the number of workers needed in December may care about the actual seasonal workload. An analyst asking whether hiring strengthened beyond the usual December increase may prefer an adjusted series.
A fictional holiday example
Imagine a shop with 100 workers in November and 120 in December, a 20% increase. Suppose its usual seasonal pattern would have implied 125 December workers under comparable conditions. The shop grew in raw headcount but did less hiring than that simple seasonal benchmark.
That benchmark is deliberately simplified. Official statistical adjustment uses time-series methods rather than assuming one fixed expected level. The example explains why “employment rose” and “employment weakened relative to seasonal expectations” need not contradict each other.
Comparisons that stay consistent
Label the adjustment status beside the number, not only in a distant footnote. For a month-to-month comparison, use the same series and units at both dates. Check whether a release reports a monthly change, a year-over-year change, or an annualized rate.
BLS periodically updates seasonal factors, which can revise past adjusted observations. Save the release date or data vintage when an exact historical comparison matters. A changed historical figure can reflect a revision rather than an error in the earlier reporting.
A practical comparison
| Reader question | Useful starting point |
|---|---|
| How many staff were actually on payroll? | Observed count with its definition |
| Did hiring outperform the recurring pattern? | Appropriate adjusted series |
| Did a published monthly change get revised? | Same series, latest vintage |
Before using the result
- Keep adjustment status consistent.
- Do not call an adjusted figure the raw count.
- Check revisions before comparing old screenshots with new tables.
Continue exploring: our economic indicators coverage.
Sources and methodology
Reviewed October 17, 2026. WorldLensia explanatory analysis. Worked examples and the accompanying teaching figure are illustrative, not measured findings or product tests. Consult the linked source for definitions and revisions.
To keep calendar adjustment separate from revisions and annualization, see Why GDP Estimates Change: Read Releases as Data Vintages and Annualized GDP Growth: A Quarterly Pace Is Not a Yearly Forecast.