How to review last month’s rosters to plan next month
A weekly check catches the problems in one roster. A monthly review shows the patterns: the day that always runs over, the shift nobody wants, the hours that creep up. Here is what to gather, what to compare, what the differences can mean, and how to carry what you learn into next month’s roster.
Most rostering happens a week at a time, and a week is too short to show a pattern. One Saturday that ran over is bad luck. Four in a row is a roster that is wrong for Saturdays. A month of rosters and timesheets side by side shows which is which, and an hour set aside once a month is enough to look.
This review sits on top of the weekly check in comparing rostered labour costs with actual labour costs. The weekly check fixes the week. The monthly review changes the plan.
What to gather
- The rosters as published for each week of the month, not the versions edited afterwards.
- Approved timesheets for the same weeks, with the actual start and finish times.
- Sales by day, and by hour if your till reports it.
- Labour cost for each week, from payroll or from the wage figures your system reports.
- A log of changes: sick calls, no shows, swaps, open shifts and who filled them.
- Next month’s calendar: public holidays, school holidays, events, leave already approved, and anyone starting or leaving.
Keep the old rosters, because the review needs them. Under the General Retail Industry Award the employer must keep a copy of each completed roster for at least 12 months (clause 15.9(c)), and the Fair Work Ombudsman’s record-keeping page says time and wages records have to be kept for 7 years.
Six measures to compare
| Measure | How to work it out | What it can show |
|---|---|---|
| Rostered and actual hours | Actual hours minus rostered hours, by week and by day of the week | Days that regularly run over or finish early |
| Labour cost percentage | Labour cost divided by sales, times 100, for each week and the month | Whether hours kept pace with trade |
| Hours on the expensive days | Hours worked on weekends and public holidays, as a share of all hours | Where penalty rates add most to the bill |
| Unplanned changes | Sick calls, no shows and open shifts, by day and by shift | Shifts that are hard to keep filled |
| Extra hours by person | Hours worked beyond each person’s rostered or agreed hours | People carrying too much, or a roster that is too thin |
| Share of the hard shifts | Weekends, closes and early starts worked, by person | Whether the least popular shifts are shared fairly |
The rostered vs actual hours calculator and the labour cost percentage calculator do the arithmetic, and the weekend hours calculator separates weekend and public holiday hours from the rest. What is a good labour cost percentage for a cafe explains what the percentage can and cannot tell you.
Read the differences before acting on them
A number on its own does not say what to change. Look for the reason behind each one:
- Actual hours above the roster on the same day every week. The day may be under-rostered, or closing may take longer than the roster allows. Check the finish times before adding a shift.
- Labour cost percentage up while hours stay flat. Sales fell, or more hours landed on higher-cost days. Separate hours, rates and sales before deciding what to change.
- Open shifts gathering on one shift. It may clash with most of the team’s availability, or it may be too short or too early for anyone to want it.
- One person with most of the extra hours. They may be the only one trained for the job. Preventing accidental overtime covers the checks.
- The same people on every weekend or close. The rotation has drifted. Fair weekend and closing-shift rotations shows how to reset it.
- Part-time staff regularly working more than their guaranteed hours. Under the Restaurant and Hospitality awards, a part-time employee who has regularly worked more ordinary hours than their guaranteed hours for at least 12 months may ask in writing for an increase (clause 10.8 of each). Under the Retail award, the employer must respond to such a request in writing within 21 days and may refuse only on reasonable grounds (clause 10.11).
When it is not clear whether a day had too many people or too few, how overstaffing and understaffing affect profit sets out how to tell. For a business with more than one site, run the review for each venue as well as the total, as tracking labour cost across multiple venues explains.
Turn the review into next month’s plan
- Fix the roster, not the timesheetWhere a day runs over every week, change the plan for that day: an extra shift, a different finish time, or less work in the close. Correcting the timesheet every week leaves the roster wrong.
- Mark the calendar firstPut public holidays, school holidays, events and approved leave on next month before placing any shifts. The public holidays page lists the dates for each state.
- Set a budget for each weekUse last month’s sales and labour cost percentage to set a wage budget for each week of next month. Forecasting staffing from weekly sales and building a roster around a weekly wage budget show how.
- Talk to the people affectedUnder the Retail, Restaurant and Hospitality awards, before changing an employee’s regular roster or ordinary hours you provide information about the change, invite their views on its impact, including on family or caring responsibilities, and consider them, unless their hours are irregular, sporadic or unpredictable (clauses 35, 33 and 39).
- Write down what to check next monthNote the one or two things you changed and the number that should move, so next month’s review starts with them.
A worked example: reviewing November at a cafe
This is an example, not a real business, and every figure in it is an example. A cafe covered by the Restaurant Industry Award reviews November to plan December. Labour cost here is the cafe’s own weekly wage figure, and the percentage is labour cost divided by sales.
| Week | Rostered hours | Actual hours | Difference | Sales | Labour cost | Labour cost % |
|---|---|---|---|---|---|---|
| 1 | 205 | 209 | +4 | $25,000 | $7,300 | 29.2% |
| 2 | 205 | 212 | +7 | $24,000 | $7,420 | 30.9% |
| 3 | 210 | 219 | +9 | $25,500 | $7,700 | 30.2% |
| 4 | 200 | 221 | +21 | $24,500 | $7,960 | 32.5% |
| Month | 820 | 861 | +41 | $99,000 | $30,380 | 30.7% |
Across the month, staff worked 41 hours more than rostered, 5% over the roster (41 ÷ 820 = 0.05). Sorting the extra hours by day put 24 of the 41 on Saturdays, 6 each Saturday: the two part-time closers each finished about an hour late every week, and a casual was kept on about 4 hours past their rostered finish to help through the afternoon. Another 11 hours came from two sick calls in week 4, covered by longer shifts than the ones they replaced, and the last 6 were spread across weekdays.
The labour cost percentage rose from 29.2% in week 1 to 32.5% in week 4, while sales stayed between $24,000 and $25,500 a week, so the rise came from hours rather than a fall in trade. The log of changes showed 6 open shifts in the month, 4 of them the 6:30 Sunday open.
What changes for December
- A 4 hour Saturday afternoon shift, 12:00 to 16:00, goes on the roster, so the extra help is planned and costed before the day rather than kept back on it.
- The closing checklist is reviewed to find out why the Saturday close runs an hour over, before anything changes for the closers. If their regular finish time does change, the owner consults both of them first (clause 33) and changes the roster by agreement or with at least 7 days’ notice (clause 15.3(d)).
- The 6:30 Sunday open is offered as a regular shift to the staff whose availability covers it, instead of being filled as an open shift most weeks.
- A second person is trained to open, so one sick call does not leave the Sunday open empty.
- December’s calendar goes on before any shifts: public holidays, school holidays, approved leave and the trading plan for Christmas.
The note for next month is short: did the Saturday difference close, and did Sunday stop appearing as an open shift? If the extra hours simply move to another day, that day becomes the next thing to look at.
Make the review a fixed date
Put the review in the diary for the same day each month, after the last week of the month has been approved and before next month’s first roster is published. An hour is enough when the rosters, timesheets and sales are ready to open, and far less is needed when they are kept together as the month goes. Reducing roster admin time each week covers keeping those records in order week by week.
Roster Flow covers the weeks the review looks back on: availability and leave from staff phones, the roster published to everybody at once, clock ins that become timesheets, and approved hours sent to Xero, MYOB or a CSV file.
A monthly roster review checklist
- Published rosters, approved timesheets, sales and labour cost for the month are gathered
- Rostered and actual hours are compared by week and by day of the week
- The labour cost percentage is worked out for each week and for the month
- Open shifts, sick calls and extra hours are counted by shift and by person
- Each difference has a reason before anything changes
- Part-time hours worked are compared with the hours agreed
- Next month’s calendar is marked before shifts are placed
- Changes to regular rosters are consulted on and given the notice your award requires
- One or two things to check next month are written down
Questions
What should a monthly roster review include?
Published rosters, approved timesheets, sales and labour cost for the month, and a log of sick calls, open shifts and swaps. Compare rostered and actual hours, labour cost percentage, weekend hours, unplanned changes, extra hours and who worked the hard shifts.
Why review monthly if I already check each week?
A week is too short to tell bad luck from a pattern. Four weeks side by side show which days and shifts are wrong every time.
What labour cost percentage should a monthly review aim for?
It depends on the business, its prices and its trading hours, so compare your own months with each other first. The figures in this article’s example are illustrations, not a benchmark.
How long do I have to keep old rosters?
Under the General Retail Industry Award, a copy of each completed roster must be kept for at least 12 months (clause 15.9(c)). The Fair Work Ombudsman says time and wages records have to be kept for 7 years.
Can I change staff rosters because of what the review shows?
Yes, but a change to someone’s regular roster brings award rules with it. Under the Retail, Restaurant and Hospitality awards you consult the employee first, unless their hours are irregular, sporadic or unpredictable, and each award has its own notice rules.
When is the best time to do the review?
After the month’s hours are approved and before next month’s first roster is published, on the same day each month.
Sources
- Fair Work Ombudsman: record-keeping
- Fair Work Commission: General Retail Industry Award 2020 [MA000004]
- Fair Work Commission: Restaurant Industry Award 2020 [MA000119]
- Fair Work Commission: Hospitality Industry (General) Award 2020 [MA000009]