Most organisations trying to make a hybrid office work are arguing from the wrong evidence. Badge swipes say who entered the building. They don't say where people sat, which rooms were booked but empty, which floor was deserted by two o'clock, or why the same six desks are always taken while thirty sit untouched. A space utilisation study replaces that argument with measurement.
How do you measure office utilisation before a refurbishment?
Measure before the design is fixed, over about a fortnight of normal working weeks. Combine the data you already hold, such as badge records and room bookings, with structured observation and a short staff survey, then compare the busiest normal day with the average. The gap between those two numbers decides how much space you can safely give up, and the rest of this article explains how to find it.
Utilisation and occupancy: the difference
The two words are often used as if they mean the same thing. They don't.
- Occupancy is how many people are in a building or a space at a given moment, compared with how many it was designed for.
- Utilisation is how well each space is used across the working day: how often a desk is actually in use, and how many of a meeting room's seats are filled when it is.
A building can have high occupancy on a Tuesday and poor utilisation all week. Both numbers matter, and a study should report each of them.
What gets measured
A proper study is observation over representative weeks, not a snapshot on a quiet Friday. It typically combines several sources:
Occupancy observation. Structured walk-throughs recording actual use of desks, meeting rooms, collaboration areas and quiet space at intervals through the day. Not what the booking system says, but what is physically happening.
Sensor and system data. If you already have desk sensors, room panels or Wi-Fi analytics, the study uses them before recommending you buy more. Most organisations own more data than they analyse.
Employee input. Surveys and interviews that ask how people actually work and what stops them. The gap between the observed data and what people report is often where the real finding sits: a floor can be technically half-empty and still feel impossible to concentrate on.
Benchmarking. How your ratios, such as desks per person, meeting seats per person and space per workstation, compare against organisations of similar size and sector.
The measures that matter
- Desk utilisation: the share of observations in which a desk was in use.
- Peak and average occupancy: the busiest normal day set against a typical one.
- Meeting room utilisation: how often each room is in use, and how many of its seats are filled.
- Booked against used: rooms and desks that were reserved but stayed empty.
- Demand for each setting: quiet rooms, phone booths and collaboration space, which are often the first to run out.
The two calculations most decisions rest on are simple:
- Desk utilisation (%) = observations where a desk was in use ÷ total desk observations × 100
- Peak occupancy (%) = the most desks in use at any one observation ÷ total desks × 100
An illustrative example: a floor of 100 desks is checked once an hour, eight times a day, for two working weeks, which gives 8,000 desk observations. If desks were in use in 3,200 of them, average utilisation is 40%. If the busiest single check, on a Tuesday at 11am, found 72 desks in use, peak occupancy is 72%. The average suggests the floor could lose more than half its desks. The peak says around a quarter at most, and fewer if teams need to sit together. That gap is why we size on the busiest normal day, not the average.
How a study runs
- Agree the questions. Which buildings, floors and teams, and which decisions the findings need to support.
- Gather the data you already hold. Badge records, room bookings, network logins and any sensors.
- Observe. About a fortnight of structured walk-throughs in normal working weeks, avoiding holidays and unusually quiet periods.
- Ask people. A short survey and a handful of interviews about how teams work and what gets in the way.
- Analyse. Peak against average for each type of space, compared with similar organisations.
- Write the brief. How much space is needed, the mix of settings, and what existing furniture is worth keeping, before any design starts.
What the numbers typically show
Utilisation studies across UK offices keep finding the same patterns. Peak desk occupancy well below what the floor plan was designed for. Meeting rooms sized for eight, used by two or three. Collaboration space that was added because it was fashionable, sitting empty while people fight over quiet rooms. And a Tuesday-to-Thursday peak that the space was never planned around.
None of these is a criticism of anyone. Desk ratios set in 2019 were set for a different way of working, and space is usually the second-largest cost on the books after people. Carrying 2019 assumptions at current rents is simply expensive.
What you do with it
The output should be a plan you can defend to a board: which teams need what, how much space that genuinely requires, what can be reconfigured rather than rebuilt, and what should not be bought at all. Sometimes the honest recommendation is to buy less than you were planning to: fewer desks, a smaller refurbishment, or space handed back at the next lease event. A study that costs a few thousand pounds routinely changes decisions worth hundreds of thousands.
It also changes the quality of the next decision. Post-occupancy evaluation, measuring again after the change, tells you whether the redesign did what it promised before the pattern gets repeated on the next floor.
When to commission one
Three moments make the measurement worth most: ahead of a lease event, when the option to shrink, grow or renegotiate is live; before a refurbishment, when the layout is about to be fixed for a decade; and when a hybrid policy has changed but the building hasn't. In each case the study needs to happen before the design work starts, because once a specification exists, the decisions that matter have already been made.
Questions we are often asked
How long should an office utilisation study run?
Where you already hold usable data, much of the analysis can start from it. Where the use of the space needs observing, allow about a fortnight of normal working weeks. That is long enough to see the weekly pattern, including the Tuesday-to-Thursday peak, without delaying the design programme.
What data should I use before a refurbishment?
Start with what you have: badge records, room booking data, network logins and any desk or room sensors. Add structured observation to show how spaces are actually used, and a staff survey to explain why. No single source is enough on its own. Badge data shows who came in, bookings show intent rather than use, and a survey on its own records opinion rather than behaviour.
Where to start
At JPA Workspaces we run utilisation studies as a fixed-price workplace diagnostic, and the findings are yours whether or not you do anything further with us. It is usually the first stage of our workplace strategy work. If you're weighing up a space decision this year, measuring first is the cheapest part of the whole project.
For the wider argument, including five questions worth asking at board level, download our four-minute briefing, The 10% that decides the 90%.