Fair workweek laws, also called predictive scheduling laws, exist because unpredictable hours are expensive for the people working them. They're local rather than national, the details differ by city and state, and the language they're drafted in isn't the language a schedule is built in. Underneath the drafting, though, most of them ask for the same handful of things.
Advance notice
The core requirement. The schedule has to be published a set number of days ahead, commonly somewhere between one and two weeks depending on the jurisdiction, and it has to reach the employee rather than be pinned somewhere they might see it. In practice this is less a legal constraint than an operational one: you can't publish a fortnight ahead if you don't know your coverage requirements a fortnight ahead.
Predictability pay
Once a schedule is published, changing it costs something. Employer-initiated changes inside the notice window typically attract extra pay, with the amount depending on how late the change is and whether the shift was shortened, cancelled or added. Changes the employee asks for usually don't attract it, and neither do genuine emergencies, though the definition of an emergency is narrower than most managers assume.
The practical consequence is that a published schedule becomes a commitment with a price attached, and the price is what makes the notice requirement real.
Rest between shifts
Most of these laws restrict the closing shift followed by the opening shift, sometimes called a clopening, either by requiring a minimum gap of around ten or eleven hours or by requiring premium pay when the gap is shorter. This one is easy to breach by accident, because each shift looks perfectly reasonable on its own and only the pair is a problem.
Access to hours
Before hiring somebody new, existing part-time staff generally have to be offered the additional hours first. The intent is to stop an employer holding a large pool of people on minimal hours. It's a rule about how you fill a gap, which makes it easy to miss when the gap gets filled quickly and informally.
Good-faith estimates, and the records underneath
New hires are usually entitled to a written estimate of the hours they can expect. And underneath all of the above sits the part that decides whether any of it is defensible.
- What the schedule said at the moment it was published
- Every change afterwards, with who made it and when
- Whether the employee requested the change or the employer did
- Any premium paid, and the reason it was owed
That's less a reporting requirement than a description of the evidence you'd need if asked. A schedule that only shows its current state can't answer any of those questions, because the version that was promised has been overwritten by the version that happened.
What to do about it
Start by establishing which rules actually apply to each of your sites, since this is local law and a chain can sit under several regimes at once. Then make notice and rest constraints part of how the schedule gets built rather than something checked afterwards. A rule that runs while you draft prevents the breach. A rule that runs at period close only tells you about it.
“The requirement underneath all of these laws is simple: mean it when you publish it.”