Formula to estimate the monthly capacity of the support team based on business days, shifts and absenteeism

Estimating an operator's daily capacity is a short calculation: messages per hour, times shift hours, minus breaks. The problem appears when that number is multiplied by 30 and used to plan the month. A month doesn't have 30 working days, the team doesn't always work with the same staffing level, and people are absent. In most of the planning workshops we run with customer service teams, the gap between theoretical and real capacity appears when the factors are reviewed one by one: it's not a universal percentage, it's what remains after subtracting non-working days, turnover, and absenteeism specific to each operation.

This page doesn't repeat the daily formula: it takes it as an input and builds the monthly formula. The result is a number of your own —the real capacity for the month— which is then compared against the volume of incoming conversations. If you don't have that second number, the formula is still useful: it tells you how much you can promise before hiring.

The four factors that turn daily capacity into monthly capacity

Daily capacity is defined as the messages an operator can answer in a full shift, once breaks and non-answering time (reading context, looking up information, escalating) have been subtracted. That number alone isn't useful for the month. To turn it into a monthly figure, it must be multiplied by four factors that are almost never 1.

• Actual working days: these are not calendar days or weekdays. They are the days the business actually operates, minus holidays and days when the channel is closed.
• Staffing factor: if the team rotates, if there are part-time staff, or if shifts don't cover every day, the average staffing level for the month is lower than the headcount.
• Absenteeism factor: vacation, leave, illness, training. It's measured as the proportion of planned operator-days that were ultimately not worked.
• Demand factor: not every day brings the same number of conversations. A month with a campaign or seasonal peak has days that consume more capacity than average, and that capacity isn't recovered afterward.

The formula, step by step

The complete formula is written like this: monthly capacity = daily capacity per operator × working days in the month × average operators per day × (1 − absenteeism) × (1 − peak reserve). Each term has an operational definition, and it's best to set them before calculating so the number can be compared month to month.

1. 1Define the daily capacity per operator with a specific shift: messages answered in a shift, not messages seen. If you don't have the measured data, use the average of a real week and not a desk estimate.
2. 2Count the business days of the month: days when the channel is open and there is at least one operator assigned. Subtract national holidays and the business's own holidays.
3. 3Calculate the average operators per day: add up the operator-days of the month (each operator who works a day counts as 1) and divide by the business days. If everyone works every day, it matches the headcount; if there is rotation or part-time, it doesn't.
4. 4Estimate absenteeism with your own data: operator-days not worked over operator-days planned in the last three months. If you don't have history, start with a conservative value and adjust it the following month.
5. 5Define the peak reserve: the percentage of capacity you want to leave free to absorb the highest-volume days. Without a reserve, any day above the average turns into a queue.
6. 6Multiply everything and compare the result with the monthly volume of incoming conversations. If capacity is lower, the difference is the deficit that must be covered with more people, more hours or less volume.

Numerical example with the formula applied

So that the formula doesn't stay abstract, it's worth seeing it with numbers. Suppose a team of three operators, each with a measured daily capacity of 60 conversations in an eight-hour shift, a month of 21 business days, with no rotation and no part-time. The theoretical capacity is 3 × 60 × 21 = 3,780 conversations. That is the number that usually appears in an optimistic spreadsheet.

Now the factors are applied. If the measured absenteeism is 8% and you want to leave a peak reserve of 15%, the real capacity is 3,780 × 0.92 × 0.85 = 2,956 conversations. The difference against the theoretical number is 824 conversations, 21.8% less. If the team also rotates and the average headcount per day is 2.7 operators instead of 3, capacity drops to 2,660. None of those adjustments is pessimism: they are conditions that already exist and that the optimistic spreadsheet was ignoring.

How to measure each factor without making up numbers

The formula is easy; the hard part is getting the inputs without overestimating. Each factor has a different source and it's best to treat them separately, because they're measured at different times of the month.

• Daily capacity: measured on conversations answered, not on messages seen. The simplest way is to count, over a normal week, the conversations each operator closed or left waiting for a customer response.
• Working days: set before the month begins, using the business calendar. Holidays are subtracted up front, not discovered halfway through the month.
• Average operators per day: calculated from the shift schedule, adding up operator-days. If there's no schedule, it's reconstructed from the channel's activity log.
• Absenteeism: measured looking back, using the last three closed months. It's the only factor that can't be set in advance with precision; that's why the historical average is used and reviewed every month.
• Peak reserve: calculated by comparing the highest-volume day of the previous month against the daily average. If the peak was 40% higher than the average, a 15% reserve isn't enough.

The three mistakes that inflate monthly capacity

Almost all capacity overestimates come from the same three places. They aren't calculation errors: they're definition errors, and that's why they repeat month after month without anyone detecting them.

• Counting calendar days instead of working days: using 30 days when the channel is open 21 inflates capacity by 43%. It's the biggest mistake and the easiest to fix.
• Ignoring rotation: if two people cover a position on alternating days, the average staffing per day is 1, not 2. A schedule that adds up people instead of operator-days doubles capacity.
• Not subtracting time that isn't answering: reading the history, looking for an answer, escalating to a colleague, entering data. That time consumes shift time and doesn't show up in the conversation count if only the moment of answering is measured.

There's a fourth mistake that doesn't inflate the number but makes it useless: calculating capacity and never comparing it against incoming volume. Capacity on its own says nothing. It says something when it's subtracted from the monthly conversation volume, because that subtraction is the deficit or the margin used to plan the next month.

What to do with the result: margin, deficit and hiring point

The final number from the formula is used in three ways. The first is to compare it against the monthly volume of conversations: if capacity is greater, there is margin and growth can be absorbed without hiring. The second is to calculate the hiring point: with the daily capacity per operator and the month's factors, you get how many additional conversations one more person covers, and it is compared against expected growth. The third is to detect the month in which the deficit becomes chronic, because a one-month deficit is covered with overtime, but three months in a row is already a staffing decision.

How this connects to customer service tools

The formula measures human capacity, and customer service tools do not change the formula: they change the value of daily capacity per operator. If the operator does not have to write every reply from scratch, their daily capacity goes up and the monthly result goes up with it. That is the only place in the formula where a tool has an impact, and it is worth measuring it before and after in the same week so that the comparison is valid. In Wando, for example, the inbox suggests replies that the team approves with one click and learns from the business's real replies; if that reduces the time per conversation, the effect shows up in the daily capacity term, not in the other factors.