Effective rostering allows a manager to respond to the unpredictable nature of customer demand and impacts profitability more directly and immediately than any other aspect of operations.

Labour cost is one of the two most expensive manageable costs in your restaurant and café, but the only one that can be managed in real-time, in response to unpredictable fluctuations in revenue.

The impact of getting a roster wrong is considerable; too many staff and the cost of sales increases, productivity drops, and service is unproductive. Too few staff result in poor customer service, lost sales, and frustrated, overworked staff.

Effective rostering is the single most proactive means of achieving sustainable profitability. It can elevate a business above the 37% of food and beverage businesses that lost money in 2017-18, or the one in five that close their doors each year.

An effective roster can help you:

1. Control labour costs
2. Meet customer demand
3. Manage staff availability
4. Meets government regulations and contractual obligations.

Controlling costs through effective rostering is a challenge; no two employees have the same skills, productivity, availability or desire to work the same number of hours, and customer demand fluctuates significantly across both the week, and seasonally throughout the year.

Is your roster performing?

A roster details when and in what role employees will work in the one to two weeks ahead. Its performance is measured through labour cost or wage cost percentage. This is calculated by adding up the wage cost of all the staff rostered that week (including salaried staff and any staff on annual leave) and dividing it by the revenue generated that week.

When we put together a roster, we calculate the value of that roster (including pension and provident fund contributions and worker’s compensations fees) and divide it by the anticipated revenue that will occur over that period. We then compare this to the actual week’s events: the revenue that was achieved, and the cost of staff servicing that revenue.

With a high-performing roster, labour costs should be:

• Close to the costs of the actual trading week’s labour cost (as a percentage of revenue)
• Close to or at the pre-defined labour cost target that has been determined to be ideal for the business to be profitable.

Here’s an example:

Steve is new to management; he manages the 100-seat fine dining restaurant in South Africa. He has 23 staff on his roster and writes his rosters two weeks in advance.

Steve takes the typical industry approach of assuming that next week’s revenue will achieve the same revenue as the corresponding week last year. The second week of August in 2018 earned R36,000, and he assumes that the same week in 2019 will do the same.

He uses an excel spreadsheet to prepare the roster, and to roughly calculate wage costs. He estimates staffing costs will be around R14,000, but doesn’t worry too much about that. He created the roster according to how many staff he needs on each shift, which is the same every week, and the other numbers usually take care of themselves.

The wage cost of last week’s payroll was 36% (including add-ons), and Steve assumes that trade will pick up in line with last year’s result of R36,000, as it did the week before. But trade doesn’t pick up. Instead, revenue actually drops. To make things worse, Steve also hasn’t accounted for pension and provident fund contributions and workers compensation in his roster, so has underestimated what the true cost will be.

As a result, the 38.9% forecast labour cost jumps to 47% of the week’s revenue. The business loses R3,480 in just one week, and Steve and his rostering process are largely to blame.

If Steve had used the previous week’s roster as a guide and maintained the previous week’s wage cost of 36%, he would have adjusted his roster to be R11,520 in value. Steve has a meeting with the owner next week to explain what happened.

What is labour cost percentage?

Labour cost percentage is the total cost of staffing your venue. It must include pension and provident fund contributions and worker’s compensation costs. It should also include accrued entitlements, such as annual leave.

You can calculate labour cost percentage as follows:

Labour during the trading week + pension and provident fund contributions + worker’s compensation fees / revenue earned in that week (less VAT)

In Steve’s case, the labour cost is R15,000 / R32,000 = 47%

To predict the cost of your upcoming roster, use a similar calculation:

Anticipated labour during the trading week + pension and provident fund contributions + worker’s compensation fees / anticipated revenue earned in that week (less VAT)

In Steve’s case, the roster cost is R14,000 / R36,000 = 38.9%

What went wrong?

When we write a roster, it becomes a mostly fixed cost in anticipation of an uncertain revenue; we simply don’t know how much money we will take each day, each service or from each guest.

On average managers spend 4% more on rosters than they should; managers and owners often under estimate the amount of revenue in the week they write the roster for and spend more on labour than they should to meet that predicted increase in demand.

A 4% saving in labour cost has a significant impact on overall profit; for a typical restaurant it would mean a 110% increase, and the typical café would see a 120% increase.

But Steve made a larger error, common to managers and owners who use spreadsheets instead of rostering software. Steve didn’t accurately cost his roster or forecast his revenue. He forgot to plan for tasks that can be controlled, such as receiving a beverage order. And finally, Steve didn’t react to changes in revenue; the roster he had written in advance was different to the one he would have written in hindsight.

He hadn’t controlled the labour cost, nor met the customer demand; fewer customers dined in his restaurant, and his labour cost blew out, resulting in thousands of dollars lost in just one week.

Effective rostering is, quite simply, the cornerstone of sustainable profitability. Conversely, ineffective rostering can undermine profitability more rapidly than any other aspect of operations.

How to manage labour costs?

Effective rostering for profit includes four stages:

1. Forecast customer demand

Forecasting demand is a cornerstone of profitable rostering. Whilst a forecast is little more than an informed guess, it is much more effective than rostering without a forecast.

Forecasting involves two stages; firstly, reflecting upon what happened 12 months ago in the corresponding week last year. Was there something different about that week? A local event that drew customers away, poor weather, a public holiday, Easter or other special event? Was that week last year significantly different from the 4 weeks before it?

Secondly, we assess the last 19 weeks of trade for the business; what is the average revenue over the last 19 weeks, per day? That is what is the average Monday trade, the average Tuesday etc?

We take the average of the last 19 weeks revenue, per day, consider if any adjustment is necessary due to special events or the weather, and write a roster to the pre-determined labour cost target.

For example, Steve now has a labour cost target of 34% (including add-ons). The average of the last 19 weeks revenue is R31,500. The labour cost target for his roster is:

• Roster Cost Target = Average 19 weeks revenue multiplied by the labour cost target
• Roster Cost Target = R31,500 x by 34 percent = R10,710

2. Determine staff availability

Define availability for staff; what days can and cannot be worked, are any staff unavailable over the rostering horizon, do staff have limitations as to the number of hours they can work, will salaried staff be rostered to work over-time, and understand regulations such as minimum duration required between two shifts.

A good rostering software package like Roubler will manage this on your behalf and provide feedback if you attempt to roster a staff member beyond their availability.

3. Schedule staff on the roster

Roster staff to a total cost no more than the target labour cost percentage value.

Great rostering software will include all the add-ons such as pension and provident fund contributions, will automatically cost in Public Holidays and penalty rates over the rostering horizon, has automatic award verification for new staff, and staff with differing experience levels and determines an exact calculation of the roster.

4. Adjust the roster to real-time information

A roster revolves around an assumed customer demand, which is unpredictable. If the predicted daily revenue is not achieved, the roster must be revised. For example, Steve’s average Monday revenue is R3,500. He has four front-of-house staff, two chefs rostered, and a roster value of R1,400.

Tuesday morning Steve realises that Monday’s revenue was only R2,500, though the wage cost was R1,375.

The weeks revenue, even if the balance of the weeks predicted revenue is correct, will be R1000 lower than the revenue the roster was written to.

Based upon the predicted R31,500 weekly Revenue, Steve had R10,710 to spend on his roster to achieve a 34% labour cost. With the loss of R1,000 Revenue on the Monday, Steve now has a roster cost of R10,370 for the whole week. He needs to find R315 savings from the next six days.

The best rostering software has automatic award recognition for new staff, and staff with differing experience, and a payroll function, which allows a manager to know exactly the cost of completed shifts, allowing the facilitating the adjustment in the roster across the balance of the week.

What is a good labour cost percentage?

Just what is considered a good labour cost percentage for a restaurant and café? The answer depends on the strategy employed by the business, and the segment it is in.

Steve manages a fine dining restaurant; it is licensed, and it is a table-service restaurant, and formal. Across its total menu, the restaurant has a food cost of 20%, and the wine list enjoys a mark-up of 220%. Table turns are slow, average spend is high, and the total business wage cost is over 30%. This is deceptive, because the kitchen wage cost is over 40%, which forces the front of house wage cost to be less than 30%.

Carmel’s café is a quick service café, where orders are taken at the counter, and menu items, like their enormous muffins, are bought in already made. Carmel’s food cost is higher, but she doesn’t have any kitchen staff, and the service required is much lower than in Steve’s restaurant. Carmel’s target labour cost is 25%.

Neither is wrong; they are pursuing the ideal wage cost percentage for their specific business in their specific segment of the industry.

24 steps to managing a roster profitably

1. 1. Utilise rostering software with built in award recognition. Excel spreadsheets cannot be easily read on smart phones, and staff often don’t have personal computers, and awards are complex and ever-changing. Also rostering software has communication channels directly to staff, and some have intuitive rostering which offers suggestions, reducing the time spent when rostering.
   2. Choose a rostering solution with integrated payroll.
   3. Integrate the rostering software with your point of sale system.
   4. Approve completed shifts in the rostering solution after every shift. Do not leave completed shifts to be approved by managers that did not work that shift, or the following day.
   5. Roster no more than two weeks in advance, adjust the second week as required.
   6. Roster start times in 15-minute windows. Instead of staff either starting at 5pm or 5:30pm, include 5:15pm.
   7. Include all add-ons in a costed roster. A roster must include pension and provident fund contributions, workers compensation and payroll tax (if applicable).
   8. Roster based upon a rand value, not based upon the number of people required each shift.
   9. Consider the weather of the upcoming week
   10. When changes are made to a roster, ensure the staff member replacing is of the same skill set and hourly rate.
   11. Schedule controllable work, such as polishing cutlery, or receiving a beverage order
   12. Align labour costs with the revenue that they generate; Don’t let the work of one shift spill into another shift, and if it does, link that cost to the previous shift. The aftermath of a lunch shift needs to be completed by the lunch staff. Staff starting an afternoon shift focus upon the tasks that relate to the afternoon shift first.
   13. Decide how to manage Annual Leave (AL) accumulation. Either the AL will be accrued in a sub-account to be paid when the staff member either takes leave, or leaves the business, or AL will be taken out of the cashflow of the business. Accruing the AL payment is preferred.
   14. Roster experienced staff on quieter nights that have variable demand. Some shifts, such as a Monday to Wednesday, can fluctuate considerably. Staffing must be orientated to the average revenue range, and experienced staff have the capability to increase productivity or complete the shift ahead of time.
   15. Measure the success of Individual days by Productivity rather than wage cost percentage. Quite often the mid-week shifts may have 40 percent or higher labour cost percentages, but their revenue per labour hour is equal to or higher than peak trading shifts.
   16. Develop and independent measure of good service, such as average wait times and maximum wait times on mains, and the time between a customer being seated, and a customer ordering for a restaurant. The number of orders per hour would apply to a QSR or café.
   17. Record the value of the managers, the front of house and kitchen staff into separate areas in the back end of the rostering platform.
   18. Ensure salaried staff are always on the roster, and clock in and out. This includes the manager.
   19. Add up the total cost of the managers salaries (including pension and provident fund contributions) and determine the daily cost of salaried staff. Salaried staff are a cost across seven days of the week, whether they are working on that day or not.
   20. Record when a staff member starts and finishes their shift independent of the rostering software. Every shift manager should have a printout in their pocket of who is rostered on, and when they start and finish. Staff can sometimes forget to clock off, or end their shift at their managers behest, only to get changed, and have a chat to a friend, and then clock off; wage theft by employees is not prevented by clock in and clock out software.
   21. Keep to the rosters cost; the roster is written based upon a specific revenue forecast and labour cost target. If the labour cost increases beyond the roster as it was written, it must be justified by the manager on duty, and should be offset by an increase in revenue.
   22. Keep labour savings out of the roster; if a day in the week performs better than expected, for example Monday’s revenue is R5,000 instead of the forecast R3,500, and the wage cost on Monday only increased by R100, a manager may think they have extra money to spend over the balance of the week. They don’t.
       Tuesday’s labour cost may well be higher as some of the clean up was left for the opening staff to do, the kitchen now needs more prep time to replace the prep that was used unexpectedly, and finally, revenue’s may drop latter in the week.
   23. Cross-train staff, so that staff members can perform a range of tasks as required.
   24. The roster is a guide; the manager on duty has the final stay when a shift ends. Some staff see a roster as a set time they start and importantly finish and can disappear when the shift is busier than expected. Make sure staff understand that the roster is a guide and confirm their finish time with the relevant manager.

Rostering is a fundamental role for a manager, and its complexity and impact on the business not only justifies but demands sophisticated rostering software with integrated payroll and automatic award recognition. Rostering software that costs a business R100 per month requires a labour saving of only four hours to make the software cost neutral. But using a powerful rostering solution like Roubler provides an opportunity to not only save time, and save money, but to actually provide the business with a tool that improves the profitability of the restaurant or café. The business can make more money when partnering with Roubler than not. And no restaurant or café can ignore that.

Words by Ivan Brewer.