How to optimize staff work schedule based on sales

When the Staff Schedule Starts Costing the Business Money

There are two equally dangerous extremes when it comes to staff scheduling. The first is having too few people on shift during the busiest hours. Queues build up at the checkout, orders take longer to complete, employees rush and make more mistakes, and some customers may simply leave rather than wait to be served. The second extreme is keeping more employees on shift than the business actually needs during quiet hours. Sales remain low while paid working hours continue to accumulate.

At first glance, the solution seems obvious: schedule more people when sales are high and fewer when they are low. In practice, that is not enough. If an owner or manager starts rebuilding the schedule every day in response to fluctuations in demand, employees end up with an unstable timetable and the business faces a new problem. Staff may be called in unexpectedly, asked to stay longer, or, on the contrary, have their shifts shortened. Such “precise adjustment” to demand does not always translate into greater efficiency.

This is clearly demonstrated by a large-scale field experiment conducted in stores belonging to the American retail chain Gap. The study was carried out by Saravanan Kesavan of the University of North Carolina at Chapel Hill, Susan Lambert of the University of Chicago, Joan Williams of the University of California, and Pradeep Pendem of the University of Oregon. The results were published in 2022 in the academic journal Management Science.

The experiment covered 28 stores in the San Francisco and Chicago areas and lasted for nine months, from November 2015 to August 2016. The researchers did not simply test whether increasing or reducing staffing levels would improve performance. Instead, they changed the scheduling approach itself: shifts became more predictable and more consistent from week to week, employees were given greater influence over their schedules, and efforts were made to provide enough working hours to those who needed them.

The outcome was especially interesting from a business management perspective. In stores that introduced more stable scheduling, productivity increased by approximately 5.1%, sales rose by 3.3%, while labour hours used declined by around 1.8%. In other words, better scheduling conditions did not mean that the company simply had to “buy” additional sales by adding more staff. On the contrary, existing labour hours were used more efficiently.

A similar conclusion emerges from a study focused directly on the restaurant industry. Masoud Kamalahmadi of the University of Miami, Qiuping Yu of the Georgia Institute of Technology, and Yong-Pin Zhou of Washington University analysed 1,444,044 transactions across 25 full-service restaurants in the United States. The study was published in November 2021, also in Management Science.

The researchers compared regular schedules with two forms of operational adjustment. The first involved notifying an employee about an additional or modified shift approximately two days in advance. The second involved changing the schedule on the actual day of work. In the first case, no significant overall decline in productivity was found. However, shifts added or changed effectively in real time were associated with an approximately 4.4% reduction in server productivity.

The researchers also identified a possible explanation: during such unexpected shifts, employees made less effort to engage in upselling and cross-selling. This is particularly important for cafés and restaurants. Staff do more than simply carry dishes or process payments. An employee may suggest a dessert with coffee, an additional drink, a larger portion, or another menu item. Therefore, a decision such as “let’s keep one more server because it might get busy” or “let’s send someone home now because there are fewer guests” affects more than just total labour hours.

This means that proper schedule optimisation is not about constantly cutting shifts or trying to keep staffing at the minimum possible level every hour. The real objective is to identify recurring periods of high and low workload, build an appropriate staffing level around them, and keep schedules as stable as the nature of the business allows.

And this type of analysis should begin not with an assumption such as “we usually get busy after lunch,” but with actual sales data.

Hourly Sales: Finding the Real Workload

The owner of a small coffee shop may be convinced that the largest customer rush happens in the morning. A store manager may believe that the most difficult period is the evening, after people finish work. Sometimes these assumptions are correct. The management problem begins when schedules continue to be built around such impressions for years without ever being checked against actual figures.

Sales patterns change. A new office or residential complex may open near the store, an educational institution may appear next to a coffee shop, customer habits may shift, and what used to be an evening peak may gradually move to lunchtime. Even within the same week, Monday and Friday may behave very differently. That is why a manager’s first task is not simply to look at total daily revenue, but to understand how sales are distributed throughout the day.

For this purpose, Kavapp Admin includes a dedicated Hourly Sales report. It allows you to see how much was sold during different time periods and to track how workload changes throughout the day. This type of breakdown is much more useful for staff scheduling than a single total figure for the entire shift.

To generate the report, an owner or manager should:
1) open Kavapp Admin → Reports → Sales Report → Hourly Sales;
2) select the required period and, if necessary, a specific sales location and cashier;
3) click Generate Report and analyse how sales change throughout the day.

However, the most important part begins after the figures are generated.

Suppose a coffee shop is open from 8:00 a.m. to 9:00 p.m. The owner believes that the greatest workload occurs between 8:00 and 10:00 a.m., so the morning shift is traditionally staffed more heavily. The hourly report may reveal a different picture: morning traffic is indeed strong, but an even more consistent peak occurs between 12:00 and 2:00 p.m., while another surge appears around 5:00 p.m. At the same time, sales drop noticeably between 10:00 and 11:00 a.m. and after 7:00 p.m.

This is no longer the manager’s impression. It is a reason to reconsider how working hours are distributed.

For example, instead of keeping the same number of employees on duty from morning until evening, shifts can be arranged to overlap so that the largest number of employees are present during the stable lunchtime peak. One employee starts earlier and finishes earlier, while another begins later, with their shifts overlapping during the busiest period. In a retail store, the same principle can be used to reinforce the checkout area during the hours when the number of sales consistently increases.

This is exactly the type of management decision for which the official Kavapp instructions recommend the Hourly Sales report: it helps identify peak hours and organise shifts so that more cashiers are working during periods of the highest workload.

However, there is one important rule: do not rebuild the schedule after a single unusual day.

If sales on Tuesday between 3:00 and 4:00 p.m. suddenly turn out to be twice as high as usual, this does not necessarily mean that an additional employee should now always be scheduled for that hour. Perhaps an event was taking place nearby, a large group of customers arrived, one customer placed an unusually large order, or a promotion was running that day.

It is much more reliable to look for a recurring pattern. When generating the report in Kavapp, the manager can change the reporting period. It is worth analysing not just one day, but several weeks, and then comparing typical days separately. If Mondays are consistently quiet while sales rise every Friday between 5:00 and 7:00 p.m., this provides a much stronger reason to use different staffing schedules on Mondays and Fridays.

For businesses with several locations, it is also important not to apply the same staffing template everywhere. The Hourly Sales report allows you to select a specific sales location, so the owner of a small chain can analyse each branch separately. A store near a business district may experience pronounced lunchtime and evening peaks, whereas a location in a residential area may be busiest in the morning and evening. Using the same staffing level at the same times in both locations may be convenient for administration, but it is not necessarily efficient for the business.

If necessary, the data can also be filtered by a specific cashier. This can be useful when a manager wants to examine a particular shift more closely or understand which sales occurred during the hours worked by a particular employee. However, this metric should not be used as a simple “cashier ranking.” Higher sales may simply mean that one employee regularly works during peak hours while another mostly works during quieter periods. This is why the workload of the location should be understood first, and only then should individual employee performance be evaluated within that context.

There is also an important limitation to keep in mind. Kavapp does not automatically create staff schedules or predict how many employees should be assigned next Tuesday at 6:00 p.m. Instead, the system provides the owner with the factual basis for that decision: data on completed sales, hourly sales dynamics, sales locations, and individual cashiers. The manager’s task is to identify recurring patterns in those figures and convert them into a practical work schedule.

This is the point at which analytics becomes a genuine management tool. You are no longer simply looking at how much the business earned during the day. You begin to understand when exactly the business needs the most hands on deck — and that is the first step towards a schedule that neither overloads the team nor forces the business to pay for unnecessary labour hours.

Not Just Revenue: Why the Number of Receipts Matters

Hourly sales show when a business earns the most. But for staff scheduling, this indicator alone is not enough. The same amount of revenue can represent completely different levels of workload for the team.

Imagine two evenings in a coffee shop. During one period, sales total UAH 20,000 and 100 receipts are issued. On another evening, revenue is also UAH 20,000, but there are 200 receipts. The financial result is the same, yet in the second case the cashier has to process roughly twice as many transactions, while the staff must handle and complete far more individual orders.

That is why, when planning shifts, it is worth looking not only at revenue but also at the number of receipts. This figure provides a clearer picture of customer flow and the workload on the checkout area and staff.

In Kavapp Admin, you can use the Receipts Report for this purpose. To generate it, go to Reports → Sales Report → Receipts Report, select the required period and, if necessary, a specific sales location and cashier. The report can also be filtered by payment method, receipt type and fiscalisation status.

If your goal is to assess workload for future scheduling, the first thing to focus on is the number of completed sales receipts during a typical working period. For example, the owner may see in the hourly sales report that revenue from 5:00 to 7:00 p.m. is only slightly higher than from 2:00 to 4:00 p.m. At first glance, there seems to be no reason to reinforce the evening shift. However, the receipts report may show that in the evening, a similar amount of revenue is generated through a much larger number of purchases.

For a retail store, this means more frequent item scanning, payment processing and receipt issuance. For a coffee shop, it means more separate orders to take, prepare and hand over. In other words, the number of transactions may explain why employees struggle to keep up in the evening even though total revenue does not look exceptionally high.

You can also calculate the average receipt value manually by dividing total sales by the number of receipts. Kavapp does not calculate this indicator automatically, but the calculation itself is very simple.

For example:
UAH 20,000 ÷ 100 receipts = UAH 200 average receipt value;
UAH 20,000 ÷ 200 receipts = UAH 100 average receipt value.

For staff scheduling, however, the average receipt value itself is less important than the reason behind the difference. If revenue remains roughly the same while the number of receipts increases significantly, the business is serving more individual purchases. That is a good reason to check whether there are enough employees working during that period.

At the same time, the number of receipts should not be turned into a single formula for calculating staffing needs. One hundred simple sales of bottled water and one hundred café orders containing several dishes create very different workloads. That is why it is best to evaluate hourly sales, the number of receipts and your own observations of how the location operates together.

This gradually shifts the question from “How much did we sell?” to a much more useful one for scheduling: “How much work was actually required to generate those sales?”

Look for a Pattern, Not a Random Peak

Even accurate figures can lead to the wrong decision if they are viewed without context. One exceptionally successful evening does not automatically mean that an additional employee should now be scheduled for the same period every week.

Suppose sales in a café rise sharply on Friday between 6:00 and 8:00 p.m. The manager decides that a new evening peak has appeared and increases staffing from the following week. Later, however, it turns out that a concert was taking place nearby that evening or that one large group placed an unusually big order. The following Fridays return to normal, while the business continues paying for unnecessary labour hours.

That is why schedules should be adjusted not because of isolated spikes, but because of recurring demand patterns.

The simplest practical approach is to compare the same days of the week. Do not combine Mondays, Fridays and weekends into one conclusion. Open the Hourly Sales report in Kavapp and review, for example, several recent Mondays. Then analyse several Fridays or Saturdays separately. If sales increase at roughly the same time on most Fridays, this already looks much more like a stable demand pattern.

Next, verify the picture using receipt data. If the number of transactions also increases regularly during the same hours, the case for reinforcing the shift becomes stronger.

In practice, a manager can follow this sequence:

1. Choose a typical day of the week, for example Friday, and review its results over the past several weeks.
2. Use the Hourly Sales report to identify the hours when higher sales repeatedly occur.
3. Check the Receipts Report to see whether the same period also has a higher number of transactions.
4. Determine whether any individual peaks were caused by a promotion, holiday, large order, nearby event or another unusual circumstance.
5. Only then adjust the schedule and monitor the results over the following weeks.

For example, if sales and the number of receipts both rise noticeably between 5:00 and 7:00 p.m. for four Fridays in a row, it makes sense to arrange an overlap between two shifts or have one employee start later so that more people are working during that period. If the spike occurred only once, it is better not to turn it into a permanent scheduling rule yet.

The same principle applies to “quiet” hours. One weak morning is not a reason to shorten a shift. But if a certain period consistently shows low sales and only a small number of receipts over several weeks, it may be worth reconsidering whether the full team really needs to be present at that time.

It is also important to account for seasonality. A schedule that worked well in November may no longer be suitable in December during the holiday period, while customer flow in a café during summer may differ substantially from winter. A pattern you identify today should therefore not be treated as a permanent rule. It needs to be checked regularly against new data.

This is where regular use of Kavapp reports becomes especially useful. An owner or manager does not need to rebuild the schedule every day. A more practical approach is to review sales and receipt dynamics once a week, while making more substantial scheduling changes only when the figures show a stable trend.

As a result, the schedule is no longer based on one unusually “good Thursday” or on the subjective impression that “we are always busy in the evenings.” It is based on recurring data. And this type of pattern provides a much more reliable answer to the question of when the business genuinely needs additional staff and when an extra employee would simply become an unnecessary expense.

How Many People Should Be on a Shift

After analysing hourly sales and the number of receipts, the most practical question arises: once we know when the busiest periods occur, how many people should actually be scheduled for the shift?

There is no universal answer such as “one employee per certain amount of sales.” Two stores with the same revenue may require different staffing levels because of differences in product range, customer numbers, service format, and the amount of additional operational work. The same applies to food service businesses: one coffee shop may mainly sell ready-made pastries and drinks, while another prepares more complex orders, so the same number of receipts may create very different workloads.

That is why it is risky to begin optimisation with the question: “Who can we remove from the shift?”

Research by MIT professor Zeynep Ton illustrates this problem well. In the Harvard Business School working paper “The Effect of Labor on Profitability: The Role of Quality”, published in 2008 and revised in 2009, the researcher analysed data from a large retail chain. She found that increasing available labour was associated with higher profitability through better execution of operational processes. One of the study’s important conclusions was that an excessive managerial focus on reducing labour costs can lead to understaffing in stores and ultimately worsen financial performance.

For a small business, this means something quite simple: an employee’s paid working hour is a cost, but not having the right employee at the right moment also has a price. It is simply harder to see. It may take the form of a queue, slower service, shelves that are not replenished on time, delays in preparing orders, or an employee who is simultaneously trying to work at the checkout, restock products, and assist customers.

That is why it is often more practical to build a schedule around the principle of “core team + peak-hour reinforcement”.

First, determine how many people the location needs to operate normally even during relatively quiet periods. This is your core team. It should be sufficient not only to process sales but also to handle all essential operational tasks: preparing the workplace, making orders, stocking and replenishing products, keeping the premises organised, receiving deliveries, covering employee breaks, and performing other routine duties.

Then return to the Hourly Sales report and receipt data in Kavapp. If you have already established that workload consistently increases, for example, between 12:00 and 2:00 p.m. and again between 5:00 and 7:00 p.m. on weekdays, you do not necessarily need to increase staffing for the entire day. It may be more efficient to arrange the schedule so that shifts overlap during those peak periods.

For example, one employee can start earlier and finish earlier, while another starts later, so they work at the same time during the two busiest hours. Alternatively, an additional employee can be scheduled not for a full shift but specifically for the period when the business consistently experiences the highest workload, provided this arrangement complies with the employment terms used by your business.

This way, the schedule begins to follow the actual demand curve instead of simply dividing the day into two equal shifts.

However, changing the schedule is not the end of the process. The new model needs to be tested in practice.

Suppose two people used to work on Friday evenings, but after analysing sales you begin scheduling a third employee from 5:00 to 7:00 p.m. After several weeks, return to the Kavapp reports and compare those periods with previous ones: what happened to sales, the number of receipts, and the results of work shifts? At the same time, assess what the report itself cannot show you: have queues become shorter, are customers being served faster, and does the team now have enough time to complete supporting tasks?

If the additional employee is regularly left without enough work, you may have overestimated the size of the peak. If sales and the number of receipts remain high while employees are still working at their limit, one additional person may not be enough.

In other words, the optimal staffing level is not determined once and for all. It is a working hypothesis that should be tested against the data: identify a pattern in Kavapp → adjust the schedule → review the result → refine it if necessary.

And there is one more important point. Even if you correctly determine the number of employees required, the shift may still perform poorly if you fail to consider who exactly will be working during the most demanding hours.

Who Should Work During Peak Hours

Imagine two shifts with three employees each. On paper, they are identical. But the first team includes an experienced cashier, an employee who knows the product range well, and a newcomer. In the second team, all three employees are still learning.

The number of people is the same, but the team’s actual ability to handle a busy evening may be completely different.

That is why, after asking “How many people do we need?”, it is worth asking a second question: “Which employees do we need at this particular time?”

This is especially important during peak hours. A busy period is not the best time for all key operations to be handled simultaneously by inexperienced employees. Ideally, the shift should include at least some people who know the workflows well, can process sales quickly, and are able to help colleagues when something unexpected happens.

For this type of analysis, Kavapp Admin allows you to use not only sales reports but also work shift and cashier reports.

A good place to start is the Work Shift Report. Open Kavapp Admin → Reports → Work Shift Report, select the required period and, if necessary, a specific sales location and cashier, then click Generate Report.

This report allows you to analyse how cashiers performed during their shifts and see who worked, when they worked, and what financial results were recorded. For a manager, this is a useful way to compare the results of a particular shift with the people who were actually working during it.

For example, suppose you have already established that Friday from 5:00 to 7:00 p.m. is consistently a demanding period. You can now review several such Fridays and compare which employees were working and what the results looked like. If a particular team composition regularly handles similar customer volumes better, this is worth considering when planning future schedules.

For a more detailed analysis, Kavapp also provides Cashier Reports. In Kavapp Admin, go to Reports → Cashier Report → Comparison by Average Indicators, select the required period and sales location, and generate the report.

This report allows you to compare employee performance using sales-related indicators, including the number of items sold, the average price of sold items, the share of products sold without discounts, and other efficiency metrics. These data can help managers notice meaningful differences, identify employees who may need additional training, and recognise those who handle sales confidently.

However, it is especially important not to turn this report into a simplistic ranking of the “best” and “worst” employees.

As we have already seen in the previous sections, an employee’s result depends heavily on when they work. A cashier who usually works during the Friday evening rush will almost inevitably record more sales than a colleague who mostly works during quiet morning hours. This does not prove that the first employee is twice as effective.

That is why employees should be compared under similar conditions: the same sales location, roughly similar days and hours, and over a sufficiently long period so that one unusually good or bad day does not determine the entire assessment.

The Cashier Breaks Report can also be useful. It records when employees start and finish their breaks. For staff scheduling, this matters for a very practical reason: even if you technically have enough employees on the shift, the team’s actual available staffing level may drop at exactly the wrong moment if breaks are poorly scheduled. Breaks should therefore also be planned with hourly workload in mind, while still ensuring that employees receive the rest they are entitled to.

In practice, the optimal shift structure can be built in the following way: first use hourly sales and receipt data to determine when additional staffing is needed, then use work shift and cashier reports to see how different employees perform under comparable conditions, and only after that decide how to distribute people across the schedule.

At the same time, numbers should not replace the manager’s knowledge of the team. Kavapp can show sales results and shift performance, but it cannot decide who is best at training new employees, who remains confident during heavy customer traffic, who handles unexpected situations well, or who can effectively cover another role when necessary.

A strong peak-hour shift is therefore not simply about “having more people.” It is about having enough people with the right combination of experience and skills at exactly the time when the business is under the greatest pressure.

Why “Matching Demand Precisely” Does Not Mean Changing the Schedule Every Day

When a business owner first starts analysing hourly sales, there may be a temptation to make the schedule as precise as possible: fewer customers today — shorten a shift; higher demand expected tomorrow — urgently call in an additional employee. From a cost perspective, this logic may seem reasonable. In practice, however, constantly adjusting the schedule at the last minute can create more problems than savings.

We have already mentioned the restaurant chain study published by Masoud Kamalahmadi, Qiuping Yu, and Yong-Pin Zhou in 2021 in Management Science. Their analysis of more than 1.4 million transactions revealed an important distinction: changing a schedule approximately two days before a shift was not in itself associated with an overall decline in productivity, whereas making adjustments on the actual day of the shift was associated with an approximately 4.4% decrease in server productivity.

There is another side to the issue as well — not only productivity, but also employee well-being.

Researchers Daniel Schneider and Kristen Harknett from The Shift Project analysed data from 27,792 hourly employees working for 80 large retail and food service companies in the United States. The data were collected between June 2016 and October 2017, and the study results were published in February 2019 in the academic journal American Sociological Review.

The researchers found an association between unstable work schedules and higher psychological distress, poorer sleep quality, and lower subjective well-being. Predictability was particularly important: employees who received their schedules less than a week in advance reported greater psychological distress than those who knew their schedules at least two weeks ahead.

This research clearly demonstrates a broader principle: a flexible schedule and a chaotic schedule are not the same thing.

For a business, the practical conclusion is straightforward. Kavapp data should be used not to create an entirely new schedule every evening for the following day, but to gradually identify stable patterns and incorporate them into schedules in advance.

If the Hourly Sales report shows over several weeks that workload increases every Friday from 5:00 to 7:00 p.m., this can already be taken into account when preparing the next schedule. If Saturdays after 7:00 p.m. are consistently quieter, that may also justify changing the distribution of working hours. By contrast, a one-off sales spike on a Wednesday should not automatically lead to permanently increasing staffing every Wednesday.

A useful principle is: build predictable fluctuations in demand into the schedule in advance, and treat unpredictable ones as exceptions.

This approach is also consistent with the Gap experiment discussed earlier in the article. Researchers Saravanan Kesavan, Susan Lambert, Joan Williams, and Pradeep Pendem examined not simply an increase in labour hours, but more stable and predictable scheduling. The results, published in 2022 in Management Science, showed increases in both productivity and sales.

So the goal of optimisation is not to create a schedule that reacts instantly to every movement in sales. It is far more important to create a schedule that is accurate enough while remaining predictable and reflecting the normal patterns of your business.

Check Whether the New Schedule Is Actually Better

Suppose you have analysed several weeks of data, identified stable peak hours, and changed the schedule. For example, three employees now work on Fridays from 5:00 to 7:00 p.m. instead of two, while during a quiet morning period you have adjusted the starting time of one employee.

Optimisation does not end there. A schedule that seemed logical based on historical data must be tested against new data. Otherwise, the business risks simply replacing one assumption with another.

Ideally, establish a baseline before changing the schedule. Take several typical weeks and review hourly sales, the number of receipts, and shift results during the periods you intend to reorganise. Then allow the new schedule to operate for some time and compare similar days and hours.

There is little value in comparing an ordinary Friday in March with a Saturday immediately before a major holiday. Conditions should be as similar as possible: the same sales location, the same days of the week, similar hours, and no obvious events that dramatically altered demand.

In Kavapp Admin, several reports can be used consecutively for this type of evaluation.

First, return to Reports → Sales Report → Hourly Sales. Compare the hours for which you changed the team composition. Has demand remained at the same level? Has the amount of sales changed since more employees were scheduled during peak hours?

Next, review the Receipts Report. If sales increased, it is important to understand what actually happened: did the number of purchases increase, or did a few customers simply place more expensive orders? This distinction matters when assessing staff workload.

Then move to the Work Shift Report. It allows you to compare the results of particular shifts, see who was working, and analyse financial transactions during cashiers’ working hours. If you are evaluating a new peak-hour team composition, compare it primarily with previous shifts that operated under similar conditions.

Finally, it is worth checking how much the new work arrangement costs the business. If cashier payroll is calculated in Kavapp, go to Reports → Financial Reports → Accrued Salaries to review payroll amounts for the required period, sales location, or individual cashier. The system takes into account hours worked and the configured components of remuneration.

This helps avoid a very common mistake: evaluating a new schedule only by looking at sales.

For example, sales may increase by 5% after a third employee is added during the evening peak. On its own, this looks positive. But if the cost of additional labour hours rises disproportionately while queues and service speed barely change, the extra staffing may have been excessive.

The opposite situation is also possible. The additional employee increases costs, but the team is now able to process significantly more receipts without delays, maintain order, replenish products, or prepare orders on time, while sales gradually increase. In that case, higher payroll costs alone do not mean that the schedule has become worse.

If desired, the owner can also calculate a simple indicator manually — sales per staff hour worked. To do this, divide total sales for the selected period by the team’s total number of working hours.

Suppose the team previously worked a combined 30 hours and generated UAH 30,000 in sales. That equals UAH 1,000 in sales per working hour. After the schedule is changed, the team works 32 hours and generates UAH 36,000 in sales — UAH 1,125 per hour. This result already gives some grounds to assume that the additional labour capacity is being used effectively.

However, this indicator should not become an end in itself either. Optimisation should not mean that every minute of an employee’s time must directly generate a sale. Staff also need time for preparation, cleaning, stocking shelves, receiving deliveries, breaks, and other responsibilities.

So after changing the schedule, do not ask only, “Did sales increase?” Ask several questions: what happened to sales, the number of receipts, team workload, labour costs, and the overall quality of operations at the location?

Only by considering these indicators together can you understand whether the new schedule is actually better.

The Schedule as a Living Model of the Business

Even a well-optimised schedule should not be treated as a finished system that can be saved and used unchanged for years. A business changes along with its customers. Tourist season begins, a new office opens nearby, the academic year changes, colder weather arrives, an outdoor seating area opens, a new promotion launches, or the product range changes. A period that was quiet six months ago may be a peak period today. That is why the best schedule is not one that you calculated perfectly once. It is a schedule that is regularly checked against actual sales patterns.

This does not mean spending hours analysing tables every day. A more practical approach is to turn report analysis into a short, regular management habit.

For example, once a week, the owner or manager can open Kavapp Admin and review several key indicators: have hourly sales changed, has a new stable peak appeared, what is happening with the number of receipts, and how have shifts performed? If there are no significant changes, the schedule can remain untouched. If a certain trend repeats for several weeks, it makes sense to investigate it more closely.

It is also worth periodically reviewing Accrued Salaries and the Profit Report in Kavapp’s financial reports. This helps avoid a situation where the schedule handles customer flow perfectly, but labour costs begin to rise faster than the financial result.

For businesses with several locations, this review should be carried out separately for each one. There is little sense in forcing all stores or coffee shops to follow the same staffing template simply because it makes scheduling easier. Kavapp allows many reports to be filtered by sales location, so decisions can be based on the actual situation at each branch.

And there is no need to change the schedule after every report review. This is precisely the difference between data-driven management and constantly reacting to numbers.

Data are not there to make you interfere with operations continuously. Their purpose is to help you notice when the old model no longer reflects reality.

If you see the same new peak for several weeks in a row, review the schedule. If the season has ended and customer flow has shifted, adapt the shifts. If results did not improve after adding another employee, check whether that reinforcement is really necessary. If a particular team performs well under heavy workload, take this into account when planning future schedules.

This is the essence of optimising staff schedules based on sales: not cutting as many working hours as possible and not scheduling as many people as possible during peak periods, but finding the right balance between demand, the team’s capacity, and business costs.

Kavapp does not create the schedule for the owner or automatically predict how many employees will be needed. But the system provides what is essential to prevent such decisions from becoming guesswork: actual hourly sales, receipt data, work shift information, employee performance indicators, and financial results. From there, a simple management logic applies: review the numbers → identify a pattern → adjust the schedule → evaluate the result → keep what works.

This is how a schedule stops being a table prepared “the way we have always done it” and becomes a fully fledged business management tool.