The weekly payroll sheet, generated instead of retyped.

Hours and overtime come out of the time clock — paired punches per store, so a person who worked two locations gets a row under each rather than one blended number. Salaried rows work the other way round: their hours are that week's published hours from the schedule at that store, which is what a salary sheet expects.

Leave gets its own columns: vacation, sick, holiday, bereavement, paternity/maternity, and an “other” column with a written reason that lands in the comments. Every one of them can be left blank for hand-fill, exactly like the paper version, or entered in the portal per employee, store and week and printed straight into the workbook. Most operators do both — the predictable stuff typed in, the odd Tuesday written on the printout.

Alongside the hours sits everything that makes a payroll sheet a payroll sheet rather than a time report: per-store salary splits, mileage, loan balances, deductions, direct-deposit notes, and free-text comments — all stored per employee and per store, because a salaried person working under three entities gets paid differently by each of them. A trailing-30-day new-hires section runs at the bottom, so onboarding paperwork and first-week pay get looked at in the same place.

And when something changed that week, the whole row turns yellow: a pay change applied since the start of the week, a note on someone's profile, or any leave hours entered. That highlight is the single most-used feature in the file. A bookkeeper scanning forty rows does not need to compare against last week — the rows that moved are the yellow ones.

Multi-entity businesses argue about this, so both layouts ship and it's one setting. By entity gives each LLC its own section with its own subtotals — the shape you want when each entity files separately and somebody reconciles section by section. Person-major puts every salaried person first and every hourly person after, one row per company beneath a bold name line with the entity label in the store column, and a subtotal row closing each section. Somebody who is salaried at three stores and hourly at a fourth appears in both sections, which is correct rather than a bug.

One rule keeps the file honest: a person gets a row at a store if they punched hours there or they have a payroll profile there. Salary specifically requires a profile row at that store — a salaried person with no profile is skipped rather than guessed at, so a salary can never duplicate itself into another entity's subtotal. That rule is shared with the timesheet and the schedule's salary split, so all three agree by construction instead of by coincidence.

Small thing worth mentioning because it decides whether people trust a generated file: the workbook declares its own sheet dimensions and row height, so it previews correctly in Quick Look and opens clean in Excel and Numbers. A payroll sheet that renders as three giant rows on a Mac gets deleted and retyped by hand, and then none of the above matters.