The Week That Drops Off

A rolling reference period is a window of fixed length that moves forward one week at a time. Each week, one week enters and one leaves. Almost all management attention is on the week entering — what somebody is about to work — and almost none on the week leaving, which has an equal effect on the figure.

The calculation in “The Week That Drops Off” depends on complete, consistently coded time records rather than a polished total at the end of the month. When considering Monitask resources for remote desktop monitoring software for remote desktop monitoring software, teams should verify exports, missing-time corrections and period boundaries against their own averaging method before relying on a dashboard.

This is the single most common reason an organisation is surprised by a breach. Nothing changed, nobody worked more than usual, and the number went up, because a quiet week from four months ago fell out of the back of the window.

For an independent reference relevant to “The Week That Drops Off”, consult the European Commission working-time resources. Use it to test working-time definitions, recordkeeping, access, retention and exception handling against the organisation’s real process rather than treating one software report as conclusive.

Why it is counter-intuitive

People reason about averages by thinking about what they are adding. That works for a cumulative total and not for a moving window, where the figure is a function of both ends.

The practical consequence is that a person can be driven into a breach by doing exactly what they did last week. If the week dropping off is lighter than the week being added, the average rises; if heavier, it falls. Two people working identical hours this week can be moving in opposite directions.

Reading the back of the window

The figure worth producing alongside the current average is next week's projected average, assuming the person works their rostered hours. It takes the same data and one more line of arithmetic.

That projection answers the only question a manager actually has: can I roster this person for this shift. The current average does not answer it, because the current average is about the past. On most systems the current figure is the only one available, which is why the conversation happens after the fact.

Light periods create debt

A run of quiet weeks lowers the average and creates room. That room is real and it is temporary: as those weeks roll out of the window, the room disappears whether or not anybody has used it.

Organisations that plan around seasonal peaks sometimes rely on this without naming it, and it works until the quiet period is shorter than expected. Naming it — we are using headroom that expires in six weeks — turns an implicit gamble into a decision somebody has taken.

The weeks that are not weeks

What goes into the window when somebody was on leave, off sick, or absent entirely is a technical question with large consequences, and it differs by regime. In some, certain absences are excluded and the window extends further back. In others they count as zero, which pulls the average down sharply.

Getting this wrong in the generous direction produces a figure that understates the real load, and it is the error that survives longest because nobody queries a number that says everything is fine.

What a usable report looks like

Per person: the current average, the projected average after next week as rostered, the number of weeks remaining in the window, and the value of the week about to drop off.

Four columns. The fourth is the one nobody has and the one that explains the movement. A manager who can see that a 38-hour week is leaving the window understands immediately why the headroom they thought they had is not there.

Fixed periods behave differently

Where the regime or the agreement uses successive fixed blocks rather than a rolling window, none of this applies in the same way: the average resets, and the end of a block is a cliff rather than a slope.

That produces its own pattern — heavy weeks early in a block are cheap, heavy weeks late in one are expensive — and it is gamed accordingly by anybody who understands it, which is usually nobody on the shop floor and one person in payroll. Knowing which of the two shapes you are running is the prerequisite for every other decision in this section.

The habit worth building

Look at both ends of the window every time the figure is produced. It takes no extra data and it changes what the number means from a description of the past into something that can be planned against.

The failure this prevents is specific and common: an organisation that monitors averages diligently, reports them monthly, and is still surprised three times a year, because it has been reading a moving figure as though it were a static one.

Explaining it to the person affected

Somebody told that their average has risen when they worked a normal week will reasonably assume the figure is wrong. The explanation takes one sentence and it has to be offered unprompted.

Your average covers the last seventeen weeks; a quiet week from the spring has just fallen out of that window and a busier one has come in. Said plainly, people accept it immediately. Left unsaid, the figure acquires a reputation for being arbitrary, and a report nobody trusts changes nothing however accurate it is.