What is the self-serve check-in window?

The early edge exists because people arrive before their shift and should not be made to wait for a button.

The late edge exists for a different reason. Somebody two hours past their start has still had a real morning, and pushing them into a review queue for it wastes an approver's attention on a routine event.

Neither edge changes what the record says. Checking in forty minutes early records forty minutes early. The window governs whether the card is available, not what time it writes.

Making both edges a setting is the only sound choice, since a tight commute city and a flexible desk team want very different widths.

A wide window is not a soft one. The card being open decides whether a button works. The time written decides everything else. Somebody checking in ninety minutes after their start records exactly that, and the lateness figure follows from it.

So widening the edges trades review effort against review quality. A narrow late edge sends more ordinary mornings into a queue. A wide one keeps the queue small and leaves lateness where it belongs, in the record rather than in an approval.

Pick the widths against real start times, then leave them alone. Edges that move every few weeks make past records harder to read, because nobody remembers which width applied when.

Missing the window entirely does not leave a gap. The card routes to a manual punch request, which somebody with approval rights reviews, so the day still resolves.

Set the two figures against your own start times before enabling the card. Sixty and a hundred and twenty are starting points shipped with the product, not a recommendation about your team.