How to Write a Staff Roster Without Losing Your Sunday Night

It's Sunday at 8pm. The spreadsheet is open. You've got fourteen names, a vague memory of who said they couldn't do Wednesday, and a feeling the roster will be wrong by Tuesday anyway. If that's where you are, this guide is for you.

Most rostering articles explain what a roster is. You already know. What you need is a process that gets you to a 90%-done draft in under half an hour — one you can publish, defend, and tweak. That's what this is.

Start with demand, not people

Don't open the staff list first. Open the trading pattern.

When are you actually busy? Saturday between 11am and 3pm? The school-holiday Tuesday lunch rush? The Thursday late-night trade? Map those windows before a single name goes on the page. A roster built around headcount — "I've got six casuals, let's spread them out" — is wrong before it's published, because it staffs your quiet hours the same way it staffs your peaks.

Sketch the week as demand bands first: peak, normal, quiet. Public holidays, school holidays, and any local event days get their own line. Now you know what shape the roster needs to be. Names come later.

Know which shifts are expensive

The next thing on the page should be a quiet acknowledgement of cost.

Under the General Retail Industry Award and the Hospitality Award, not every hour costs the same. Weekend trade, late nights, public holidays, and early starts carry penalty rates. Casual loading sits on top. A Sunday peak isn't just a busy shift — it's an expensive busy shift, and how you staff it should reflect that. (If you need a refresher on the multipliers, our penalty rates explainer lays them out.)

This isn't a compliance lecture. It's a pricing reality. If your most expensive trading window is being covered by your most expensive labour mix — three casuals on a Sunday afternoon when a part-timer could anchor it — you're paying twice for the same hour. Knowing which shifts cost what shapes how you staff them.

Build the skeleton with your reliable people first

Now bring in names. But not all of them.

Start with your full-timers and your regular part-timers — the people whose availability barely changes. Anchor them into the high-demand windows you mapped in step one. The Saturday peak gets your strongest closer. The Sunday open gets the person who actually shows up at 7am without a text reminder.

This is the skeleton. It usually fills 60–70% of the roster in about five minutes, because these people have predictable shifts. You're not making decisions — you're recording ones you already made when you hired them.

Then, and only then, fill the gaps with casuals. You're now choosing between three or four people for a specific slot, not staring at fourteen names and a blank week.

Rotate the bad shifts on purpose

Every venue and every store has bad shifts. The Sunday open. The Friday close. The 6am stocktake. The public-holiday late.

If the same casual closes every Friday night, they notice. They might not say anything for a month. Then they're gone, and you're wondering why a reliable person quit "out of nowhere." They didn't. You rostered them out of the place.

Fair rotation is a retention decision, not a scheduling one. Keep a mental note — or better, a visible note on the roster — of who pulled the bad shift last week. Rotate it. If someone prefers the Sunday open because it suits their week, great, lock it in. But default to rotation, not default to convenience.

Publish early, and to phones

A roster nobody has seen is the same as no roster.

If your staff are checking a group chat at 7am the day of their shift to find out when they start, you don't have a system — you have a daily emergency. The fix isn't more group chats. It's publishing the roster at least a week in advance, to a place every staff member checks without thinking: their phone, with a push notification when it changes.

This also kills the "I didn't see it" excuse. When the roster lives in one place and changes trigger a notification, the responsibility moves from you to them. That's where it should sit. (Here's how Schedaddle handles publishing and changes if you want the short version.)

Have a callout plan before the callout

The 6am Sunday text — "I'm so sick, sorry" — is going to happen. The question is whether you've planned for it or you're starting from zero with a coffee in your hand.

For every shift window, identify one or two casuals who are typically available and willing to pick up. Note them on the roster itself — a small "backup: Jess, Tom" line next to the shift. When the callout lands, you're not scrolling through fourteen contacts trying to remember who said they wanted more hours. You're sending two texts.

You won't always land it. But going from "no plan" to "two names" turns a panic into a two-minute task most weeks.

The roster is a draft, not a finished document

Here's the mindset shift that saves the Sunday night.

You are not writing a perfect, final document. You are writing a 90%-done draft that you'll tweak across the week. Someone will swap. Someone will call out. A public holiday will surprise you. That's not failure — that's rostering.

If your weekly process takes more than 30 minutes, something in the system is broken. Usually it's one of three things: you're starting from a blank page each week instead of last week's roster; you're trying to please every availability preference in your head instead of on paper; or you're the only person who knows how shifts get swapped, so every change routes through you.

The loop, not the task

Writing a roster isn't a thing you finish on Sunday. It's a loop.

Sunday: demand → cost → skeleton → fill → rotate → publish → backup plan. Monday to Saturday: handle the swaps, the callouts, the one casual who can suddenly do Thursday. Then Sunday again, but this time you start from last week's roster, not a blank one. The first week is the slow week. The fifth week, if you've got the loop right, is twenty minutes with a cup of tea.

That's the goal. Not a perfect roster. A repeatable one.

If you're stuck on the same Sunday-night loop, I'd genuinely like to hear which part of it eats your evening. The demand mapping? The callout scramble? The casual who keeps copping the Friday close? Tell me where your roster breaks — that's the only way the next version of this guide gets better.