BUSI 3014 Week 8 Seasonal Trend Read Example

Three Years of Monthly Service Requests Read for Season, Direction and Odd Months: Spring Peaks That Rise About a Tenth a Year, a Phone Outage in August 2026 and a Cold-Snap November in 2027

Student Name

Walden University

BUSI 3014: Data Driven Decision Making Workplace Certificate

Instructor Name

Month Day, Year

Three Years of Monthly Service Requests Read for Season, Direction and Odd Months: Spring Peaks That Rise About a Tenth a Year, a Phone Outage in August 2026 and a Cold-Snap November in 2027

The Series

The series is the number of new service requests received each month by the Knoxville pest control firm in this course (a composite) from January 2025 through December 2027. A request is any call or web form asking for service from a new or returning customer, so the series counts demand arriving at the dispatcher's desk, not work completed. It is drawn below as a line, one point per month.

Figure 1

New Service Requests per Month, January 2025 to December 2027

Kestrel Pest Solutions and its request counts are composite.

The Seasonal Shape

The line rises and falls on the same rhythm every year. Requests climb from about 120 a month in January and February to a peak of 310 to 390 in June and July, as ants, wasps and termites become active, then fall through the autumn to a low in December. A smaller bump appears in late fall in some years, when mice and rats move indoors. This pattern is the most obvious thing in the chart, and it is not news: it is the shape of the pest control year. Any comparison of one month with the next will mostly reflect it. Hyndman and Athanasopoulos (2021) describe seasonality as a pattern that repeats over a fixed period, and the first task in reading a series like this is to separate that repeating pattern from everything else.

Same Month Against Same Month

Setting every month beside its counterpart twelve months before is the plainest way to take the season out.

Table 1

Selected Months Compared With the Same Month a Year Earlier

Composite data. Changes are percentages of the earlier year's figure.

The Underlying Direction

Compared month for month, the whole pattern is shifting upward. Most months are 7 to 14 percent higher than a year earlier, and full-year requests rose 10.8 percent in 2026 and 12.0 percent in 2027. The peaks themselves are rising: July went from 311 to 352 to 389. A three-month moving average, which smooths out single-month noise, shows the same thing, with each year's summer high and winter low sitting above the previous year's.

The direction is upward, then, at roughly a tenth a year. That describes what happened from 2025 to 2027. It does not say what will happen in 2028, and this read does not project the line forward. Growth could slow if a competitor opens, or speed up if the company adds advertising.

Months That Do Not Fit

Two months break the pattern. August 2026 fell 3.4 percent below August 2025, while every summer month around it was up by more than 10 percent. The dispatcher's log explains it: the company moved to a new phone system that month, and calls to the main number were dropping for about nine days before the problem was found. The dip reflects missed calls, not lower demand.

November 2027 rose 27.2 percent over the year before, far above the typical increase. An early cold snap in the first week of November brought a wave of rodent calls weeks ahead of the usual timing. The spike is real demand, but it came from weather, not from a change in the business.

Both months should be noted whenever the series is used, and neither should be read as a sign that the underlying direction changed. Cleveland et al. (1990) designed a widely used method for splitting a series into seasonal, trend and remainder parts, and these two months would show up in the remainder, the part neither the season nor the trend explains.

How the Moving Average Helps

A three-month moving average, plotted over the raw line, smooths the single-month noise and makes the yearly peaks easier to compare. It also softens the two unusual months, which is useful for seeing the direction but would hide them if it were the only line drawn.

What This Means for the Saturday Route Question

The series bears on the Saturday route decision in two ways. Demand in the peak months is high and rising, so the capacity shortfall measured in the spring of 2027 is unlikely to ease on its own. And the peak runs from April to September, which matches the months the owner is considering for a Saturday route. The series does not show whether customers waited, only that more of them called; the wait itself comes from the measure defined earlier in the course. Tufte (2001) argued that good statistical graphics invite comparison across time while keeping the data honest; drawing the full three years, rather than a single season, is what makes both the rhythm and the rise visible.

References

Cleveland, R. B., Cleveland, W. S., McRae, J. E., & Terpenning, I. (1990). STL: A seasonal-trend decomposition procedure based on loess. Journal of Official Statistics, 6(1), 3-73.

Hyndman, R. J., & Athanasopoulos, G. (2021). Forecasting: Principles and practice (3rd ed.). OTexts. https://otexts.com/fpp3/

Tufte, E. R. (2001). The visual display of quantitative information (2nd ed.). Graphics Press.