The Register of Who Has Opted Out
Most regimes require a record of who has agreed. Most organisations have a drawer of signed forms, which is not the same thing and cannot answer any question.
Where an individual opt-out is permitted, the employer is usually required to keep a record of the workers who have agreed. Some regimes specify a register; others simply require the agreements to be retained and producible.
The workflow in “The Register of Who Has Opted Out” becomes more reliable when scheduled hours, actual time and later corrections can be distinguished. For teams exploring how to handle multiple clients, this workforce software resource can add operational time and project context, provided data collection is proportionate, permissions are limited and every important exception receives human review.
What exists in practice is a folder, or a sub-folder of each person's HR file, containing scanned forms. That satisfies retention and answers no operational question at all, which is why the rota system does not know who has signed.
For an independent reference relevant to “The Register of Who Has Opted Out”, consult the Atlassian project-management guide. 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.
What a register is, as opposed to a folder
A list, with a row per person, that can be read, counted, filtered and fed to another system. Name, date signed, notice period for withdrawal, current status, and the date of the last review.
Five columns. The difference between this and the folder is that the register can answer questions: how many people, which departments, who withdrew, and — most importantly — which flag should the rostering tool suppress for this person.
Why the rota tool needs it
Covered elsewhere in this collection and worth repeating here: the weekly average flag means something different for somebody who has opted out, and a system that shows a breach state for a person who has validly agreed trains everybody to ignore the state.
Feeding opt-out status into the rostering tool requires a list rather than a folder. This is the single most common reason the working time indicators in a rostering system are switched off: they were producing false positives for half the workforce.
Where it should live
In the HR system, as a field on the person record, with the signed document attached. Not as a standalone spreadsheet, because the spreadsheet will not be maintained when the person who built it leaves.
A field on the record also means it migrates, appears in exports, and can be fed downstream without anybody doing anything. A spreadsheet means a manual step that will be skipped during the month it matters.
The fields that are usually missing
The date signed. Surprisingly often absent, which makes it impossible to say when the agreement took effect.
The notice period, where it was agreed as something other than the default. And the review date, if the organisation reviews them — which it should, and which is the subject of the next note.
Without the date signed in particular, an organisation cannot establish that the agreement predates the hours it is relying on it for, and that is the exact question that gets asked.
Reconciling it
Twice a year, match the register against the current headcount. Expect three categories of discrepancy.
Leavers still on the register, which is tidy-up. People on the register with no form in the file, which is a records problem and needs re-papering. And people whose rota shows a sustained average above the limit with no entry in the register at all — which is the one that matters, because it means the organisation has been relying on an agreement that does not exist.
The third category is the point of the exercise
It is also the one nobody looks for, because it requires joining the register to the hours data rather than to the headcount.
One query: everybody whose average exceeded the limit in the last reference period, left-joined to the register. Anybody with no match is a breach that has already happened. On a first run, most organisations find a few, usually people who transferred in, or agency workers whose agreement sits with the agency, or somebody whose form was never filed.
What to do with what you find
Papering over it retrospectively is not available: an agreement signed today does not cover hours worked last month. The honest actions are to record what happened, to obtain an agreement going forward if the person wants to give one, and to adjust the rota in the meantime.
Then fix the process that produced it, which is nearly always a missing step at transfer or at the point somebody moves from agency to direct employment. That step costs a line on a checklist, and its absence is the reason the register and the reality drifted apart in the first place.
What to do about the forms with no date
Old agreements frequently have a signature and no date, or a date in a handwriting nobody can read. They cannot be fixed retrospectively: a date added now is not evidence of when it was signed.
The remedy is to re-offer, properly, to everybody in that position, and to record the new agreements with dates. It is a morning of administration and it converts a stack of documents that would not survive a question into a register that would.