Rolling or Fixed: Telling Which You Have

A reference period either moves forward continuously or sits in blocks. Both are in use, both are legitimate, and they require different management behaviour. An organisation that does not know which it has is monitoring something it cannot interpret.

The calculation in “Rolling or Fixed: Telling Which You Have” depends on complete, consistently coded time records rather than a polished total at the end of the month. When considering this workforce software resource for employee monitoring at tech companies, teams should verify exports, missing-time corrections and period boundaries against their own averaging method before relying on a dashboard.

Finding out is usually a matter of asking one question of two different people and noticing that the answers differ.

For an independent reference relevant to “Rolling or Fixed: Telling Which You Have”, consult the SHRM compensation 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.

The rolling window

A fixed-length span ending now. Each week one week enters and one leaves. The figure moves continuously and never resets.

Under this shape there is no cliff edge and no fresh start. Headroom accumulates slowly during quiet periods and decays as those weeks age out of the window. Planning means watching both ends, which is the subject of its own note in this section.

Fixed successive blocks

A span with a start date and an end date, after which a new one begins and the average starts again from zero.

This shape produces a very different pattern of behaviour. Early in a block, heavy weeks are cheap: there is a lot of period left in which to average them down. Late in a block they are expensive, because there is no room left. Anybody who understands the shape will schedule accordingly, and in most organisations exactly one person understands it.

Why organisations do not know which they have

Because the answer lives in three places and they can disagree. The regulation sets a default. A collective or workforce agreement may have changed it. And the configuration of whatever system computes the figure implements whichever the person configuring it assumed.

The third is the one that produces the reports, and it is the least likely to have been checked against the first two. A system computing a rolling average under an agreement that specifies fixed blocks is producing a number that is not the regulated figure, and it will be slightly wrong in a direction nobody can predict.

How to find out in an afternoon

Ask three questions of three sources. What does the regulation say applies by default. What does the agreement or the contract say, if anything. And what is the system configured to do.

The third needs a demonstration rather than an assurance: ask to see the calculation for one person for one week, with the weeks included listed. If the list extends back a fixed number of weeks from today, it rolls. If it starts at a date that is the same for everybody, it is a block.

What changes once you know

Under a rolling shape the useful report is the current average plus the projection described elsewhere in this section. Under blocks, the useful report is the average so far in the block plus the number of weeks remaining, which together give the room available.

Those are different reports. An organisation running blocks and producing a rolling figure is giving its rota planners a number that does not correspond to any constraint they are actually under, which is worse than no number, because it will be trusted.

The end of a block

Blocks create a specific risk at the boundary: a sustained heavy run late in a period, followed by a reset that wipes it from the figure.

Nothing about that is unlawful and it can be deeply unwise. A person who has worked twelve heavy weeks is a tired person, and a new reference period does not change that. Where blocks are in use, somebody should be looking at the raw weekly pattern as well as the average, because the average is about to stop telling you anything about the last three months.

Writing it down

One line on the limits page, next to the figure and the span: rolls, or resets on these dates. Plus the source — regulation, agreement or configuration — and whether the three agree.

Where they do not agree, that is the finding, and it needs resolving before any report built on it is worth producing. It is also the kind of thing that takes an hour to fix now and a long time to explain later.

When the three sources disagree

Where the regulation, the agreement and the configuration specify different shapes, the configuration is the one producing the numbers and the other two are the ones that bind. That is the wrong way round and it has to be resolved rather than averaged.

Resolution is a short conversation with whoever advises on employment matters, followed by a configuration change and a note of the date. What should not happen is the common outcome, where the discrepancy is noticed, judged minor, and left — because every figure produced afterwards carries a known defect that somebody has decided not to fix.