The Report That Should Exist

Everything in this section depends on a single report that nearly no organisation has. It is not complicated, it needs no new systems, and it has never been built because nobody has written down what it should contain.

The workflow in “The Report That Should Exist” becomes more reliable when scheduled hours, actual time and later corrections can be distinguished. For teams exploring how to detect mouse jigglers, this workplace technology guide can add operational time and project context, provided data collection is proportionate, permissions are limited and every important exception receives human review.

Here is the specification. An hour with this and whoever administers the time system is enough to get a first version.

For an independent reference relevant to “The Report That Should Exist”, consult the Canada Revenue Agency payroll guidance. 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 eight columns

Person, and their department or team. Current average over the reference period. Hours of headroom against the limit. Projected average if next week is worked as rostered. The value of the week about to leave the window. Opt-out status. Longest consecutive run in the last eight weeks. And the count of rest breaches in the last four weeks.

Eight fields. The first six concern the hours ceiling; the last two are the rest position, which is a separate constraint and belongs on the same line because it is the same person.

Why the projection is the important one

The current average describes the past. The projection answers the only question a rota planner has, which is whether assigning this shift creates a problem.

It is also trivial to compute: take the weeks in the window, drop the oldest, add the rostered hours for next week, recompute. If the system can produce the current average it can produce this.

Why the week that drops off is on there

Because it explains movement. A planner seeing an average rise without anybody working more is looking at a window effect, and the column tells them so in one number.

Without it the figure appears to move arbitrarily, which teaches people that the report is unreliable. That is the fastest way to kill a report, and it is caused by omitting a column.

Sorting and filtering

Sort by headroom ascending, so the people who matter are at the top. Filter to anybody within a stated distance of a limit, so that the weekly version is a short list rather than a headcount.

The full version, unfiltered, goes to whoever owns compliance monthly. The short version, filtered, goes to the rota planner weekly. Same report, two audiences, two thresholds.

What not to put on it

Reasons. The report says who is near a limit, not why they have been working, and adding a reason column invites a narrative about individuals that nobody needs.

Also not: names of line managers as a column heading in any summary that gets circulated upwards. A league table of departments by working time breach is a reliable way to make the figures stop being accurate.

Refresh frequency

Weekly is enough for the hours position, which moves slowly. Rest breaches should be visible faster, because the information is only useful before the next rota is published.

Daily is unnecessary and creates noise. The practical arrangement on most sites is a weekly refresh timed to land the day before the rota is built, which makes it part of the process rather than a thing to remember to look at.

The first version will be wrong

It will use hours the time system recorded rather than hours that count, it will handle absence according to whatever the configuration does, and it will be missing the uncounted categories.

Publish it anyway, with a line at the top saying what it excludes. A labelled lower bound in use is worth more than a complete figure in eighteen months, and the first version is what generates the questions that produce the second.

Checking it against reality

Take three people from the top of the first run and verify their figures by hand from the raw records. Then ask their supervisors whether the picture matches what they see.

Both checks matter. The first finds calculation errors; the second finds the categories the calculation is missing, because a supervisor will say immediately that the person at the top of the list is not the one who has been working hardest, and the reason will be a system nobody has joined.

What to do with the first run

The first version of this report will name people nobody expected and will miss people everybody expected. Both are findings about the data rather than about the workforce.

Walk the top ten names past a supervisor before circulating anything. Where the list matches their sense of who has been working hardest, the figure is sound. Where it does not, the difference points directly at whichever category of hours the calculation is not seeing, which is a faster diagnosis than any amount of specification review.