How it works
Peak occupancy is the maximum of an occupancy series, where occupancy is the share of available units in use. Record it by day or by hour over several normal weeks, then take the highest value, or a high percentile such as the 90th, so one unusual event does not set your capacity. It counts how many units are in use at the busiest point, not how intensively they are used across the day; that is utilization.
In hybrid offices the peak is concentrated. People come in when the colleagues they work with are in, so attendance clusters midweek, often on Tuesday and Wednesday, while Monday and Friday stay quiet. Space sized to the average fails on the busiest days; space sized to an unmanaged peak sits empty the rest of the week.
Peak occupancy feeds the desk ratio and the footprint. A team whose busiest normal day brings in 60 of its 100 people needs about 60 desks plus a buffer, whatever its average. Policy moves the peak: spreading team days across the week lowers it, and a new office-days requirement raises it.
Formula
Peak occupancy rate = (Highest units in use on any day ÷ Available units) × 100
- Highest units in use on any day
- The largest number of desks, seats or rooms in use on any single normal day (or hour) in the period, from check-ins or presence data.
- Available units
- The desks, seats or rooms that were open and bookable in the same area.
Worked example · illustrative figures
- A floor has 150 bookable desks.
- Presence data for a normal week shows 60, 117, 105, 75 and 30 desks in use from Monday to Friday.
- The busiest day is Tuesday: 117 ÷ 150 = 0.78. The average day is 77.4 ÷ 150 = 0.52.
Peak occupancy: 78% on Tuesday, against an average of 52%.
How to measure it
- Use presence data (check-ins, network presence, badge data), not bookings; cancellations and no-shows make booked peaks look higher than real ones.
- Track the peak by weekday and by hour for each floor and neighborhood, not only for the building.
- Leave out one-off events such as all-hands days, or report them separately, and use a percentile if single spikes distort the figure.
- Check again after every policy change, since anchor days and office-day requirements move the peak.
Examples
- An office averages 45% occupancy but reaches 90% every Wednesday. Planning to the average would leave people without desks one day a week.
- A council assumed Wednesday was its busiest day. Presence data showed Thursday, so it moved team days and services to match.
Benchmarks
CBRE's benchmarking reported average peak utilization (the share of seats in use at the busiest point) of 80%, above the 65% target most respondents set.
CBRE, The Hybrid Reality (2026 Global Workplace & Occupancy Insights)In the same benchmarking, about 73% of respondents said Tuesday has their highest attendance.
CBRE, The Hybrid Reality (2026 Global Workplace & Occupancy Insights)Denbighshire County Council found its real peak day was Thursday, not the Wednesday it had assumed.
Kadence customer story: Denbighshire County Council
Common mistakes
- Sizing space to the weekly average.
- Treating a one-off event day as the normal peak.
- Reading the peak from bookings; cancellations and no-shows make booked peaks higher than real ones.
- Trying to flatten the peak by asking individuals to come in on quiet days, without changing when teams coordinate.
How Kadence helps
Kadence Workplace Insights shows attendance and occupancy by day, with office heatmaps that make the busiest days and areas visible, built from check-ins and presence rather than bookings alone. Team coordination features show people who else is going in, which helps teams choose their days, and SpaceOps plans floors against the occupancy data WorkOps records.
Workplace analytics · Hybrid working coordination · SpaceOps
Frequently asked questions
What is peak occupancy in an office?
It is the busiest level of use a floor or building reaches in a normal week, usually on one or two midweek days. It matters more than the average for capacity planning, because it decides whether the people who come in can find a desk or a room.
Why are offices busiest on Tuesdays and Wednesdays?
People come in when the colleagues they work with are in, so in-person time clusters on the same midweek days. That is a rational response to hybrid work, not a policy failure. Plan the office to work well on those days, and decide separately what to do with the quiet ones.
Should you plan office space for peak or average occupancy?
Plan to a chosen peak, not the average. Use the busiest day of a normal week, or a high percentile of daily occupancy, then decide how much buffer you want and whether team scheduling can lower the peak before you commit to space.
How do you reduce peak occupancy?
Spread team days deliberately rather than asking individuals to pick quiet days. Assign neighborhoods or anchor days by team, show people who is going in, and release unused bookings so booked-but-empty desks do not make the peak look worse than it is.
What is the difference between peak occupancy and utilization?
Peak occupancy is the highest share of desks, rooms or area in use at the busiest point of a normal week. Utilization is how intensively capacity is used over time, combining how often spaces are used with how full they are. A hybrid office can hit a high midweek peak while its utilization across the week stays low.
Related terms
- Occupancy rate The share of available desks, rooms or floor area in use, at a point in time or on average over a period.
- Occupancy vs utilization Occupancy is the share of space in use; utilization is the share of capacity-time actually used.
- Space utilization How intensively buildings, floors, rooms and desks are used against their full capacity over time.
- Desk ratio (desk sharing ratio) The number of people assigned to a workplace for each desk, set from peak attendance and policy.
- Anchor days Set days each week when a team or company agrees to be in the office together, with the other days flexible.
- Office heat map A floor plan colored by how often each desk, room or zone is used, so busy and quiet areas stand out at a glance.
Sources
Kadence editorial team · Last reviewed