'We are short-staffed' and 'we have too many people' can both be true at the same time, depending on the time slot. Sizing a workforce is not an opinion: it is a calculation. The difficulty is not in the maths, but in making the right assumptions (peaks, absences, skills) and reviewing them with real data.
1) Start with the need: coverage hours by time slot
Before looking at contracts, define coverage: how many people do you need, when, and with what skills. In operations, the need changes by hour, not by day. If you only define a daily number, you hide the problem.
Example: a shop needs 2 people in the morning and 6 in the afternoon. If you plan '4 all day', you will have overcoverage in quiet periods and undercoverage during peaks, and end up paying overtime or losing sales.
2) Convert coverage into FTEs (and do not forget 'non-productive' hours)
Once you have coverage hours, convert them into full-time equivalents (FTEs). This is where breaks, pauses, meetings, training, and non-visible tasks (closings, stocktakes) come in. If you do not count them, the plan breaks down.
Example: if you need 400 weekly hours of actual coverage and each person contributes 37.5–40 hours depending on their contract, the number of FTEs is clear. What usually goes wrong is not counting breaks, sick days, or extra tasks.
3) Add a buffer for absenteeism and turnover (this is realism, not pessimism)
Every operation has a level of absences. If you size without a buffer, any sick day pushes you towards overtime. The buffer does not have to be fixed: it can vary by season or by location.
Example: if historically a location has more sick days in winter, the annual plan should reflect reinforcements or an hour bank for that period. Without that, the 'plan' only exists in a spreadsheet.
4) Quick example: a 24/7 operation
Imagine a minimum need of 3 people per shift, 3 shifts a day, 7 days. For baseline coverage alone, you already have many weekly hours. If there are also critical skills (for example, a role that cannot be left unfilled), you need multiskilling or reinforcements.
The goal is not to memorise formulas, but to understand that 24/7 requires design: rotation, rest periods, back-ups, and a system that measures extensions and changes for adjustment.
5) Win-win: fewer emergencies and better service
For the company, proper sizing reduces chronic overtime and improves quality. For the worker, it reduces overload and last-minute changes.
Data-driven sizing turns the conversation from 'gut feelings' to decisions: where to reinforce, where to redistribute, and which process is generating wasted time.