The short version
- Most offices run well below capacity. Across HybridHero platform benchmarks, average utilisation sits at 54% against a 60 to 70% target.
- You cannot cut what you cannot see. Measure real desk and room usage for at least a quarter before touching the lease.
- Attendance is lumpy. Midweek peaks near 60% while Fridays sit in the mid 30s, so size for the peak you choose to serve, not the average.
- Right-sizing decisions come in three flavours: renegotiate, sublet or consolidate. The data tells you which one fits.
- Track cost per desk and cost per attended day, not just rent per square metre.
The fastest way to reduce occupancy costs is to stop paying for space nobody uses. That sounds obvious. In practice, most firms sign leases based on headcount, then run hybrid policies that mean half the desks sit empty on any given day. The gap between what you pay for and what you use is the single biggest line of recoverable spend in most workplace budgets, and closing it starts with measurement, not with a property negotiation.
I have sat in enough lease reviews to know how these conversations usually go. Finance wants a number. Property wants certainty. Workplace teams get asked to guess. The firms that actually cut space costs do it differently: they run three to six months of real utilisation data, work out their genuine peak demand, then make one of three moves with confidence. This article walks through that process, because guessing wrong in either direction is expensive. Too much space wastes rent. Too little wrecks the employee experience and sends people straight back to working from home.
Know your real occupancy cost first
Before you can reduce anything, agree what the number actually is. Occupancy cost is not just rent. It is rent plus service charges, utilities, cleaning, security, insurance, fit-out amortisation and facilities staffing. Most organisations only track the rent line, which understates the true cost of every empty desk by a wide margin.
The more useful working metric is cost per desk: total occupancy cost divided by the number of desks you operate. It turns an abstract property figure into something you can act on. If you close a floor, you can state exactly what each removed desk saves. If you add bookable desks to replace fixed ones, you can show the ratio improving quarter on quarter.
Go one step further and divide occupancy cost by attended days rather than desks. A desk that gets used twice a week costs you the same as one used five days a week. Cost per attended day exposes that difference, and it is the number that makes boards sit up.
Measure utilisation before you touch the lease
This is the step most firms skip, and it is the reason so many downsizing projects go wrong. Badge data tells you who entered the building. It does not tell you whether they found a desk, whether the fourth floor sat empty while the second was rammed, or whether your meeting rooms were booked solid but half unused.
Booking and check-in data from a desk and room platform gives you the picture badge swipes cannot: which desks, which neighbourhoods, which rooms, on which days, by which teams. Feed that through reporting and analytics and you get the three numbers every right-sizing decision depends on: average utilisation, peak utilisation, and the day-of-week spread. Run it for a full quarter at minimum. Anything shorter and you will mistake a school holiday or a project crunch for a trend.
The benchmarks are sobering. Across 1,500+ workplace teams on the HybridHero platform, average utilisation runs at 54% against a healthy target of 60 to 70%. Meeting rooms are worse: no-show rates of 25 to 35% are normal, meaning a third of your most expensive space is booked and then abandoned.
Desk utilisation: benchmark vs healthy target
Average utilisation54%
Target range (lower)60%
Target range (upper)70%
Source: HybridHero Workplace Visibility Report benchmarks.
Size for your chosen peak, not the average
Attendance in hybrid workplaces is not flat. HybridHero platform benchmarks show midweek attendance peaking near 60% of headcount, with Fridays down in the mid 30s. If you size your office for the Tuesday to Thursday peak, you carry empty space on Mondays and Fridays. If you size for the average, midweek becomes a scramble for desks and people stop coming in.
The honest answer is that peak demand is a choice, not just a measurement. You can flatten the peak with team scheduling, anchor days spread across the week, and booking rules that nudge demand towards quieter days. Every percentage point you shave off the Tuesday peak is space you no longer need to lease. Firms that manage the curve before they cut the footprint get a smaller office and a better one. Firms that cut first and manage later get complaints.
Do not cut to the average. Size the footprint for a managed midweek peak, then use booking rules and team days to keep demand inside it. The lease saving comes from flattening the curve, not from hoping it flattens itself.
Renegotiate, sublet or consolidate: which move fits
Once you have a quarter of clean data, the decision usually resolves into one of four options. Each suits a different situation, and the table below is the framework I use when finance asks which lever to pull.
| Approach | Best suited to | What the data must show | Main risk |
|---|
| Do nothing yet | Utilisation already inside target range | Peaks regularly approaching capacity | Complacency if hybrid patterns shift |
| Renegotiate the lease | Break clause or renewal within 12 to 18 months | Sustained utilisation below target | Landlord holds the timetable, not you |
| Sublet surplus floors | Long lease, clearly separable space | Whole floors consistently underused | Sublet demand and consent from landlord |
| Consolidate sites | Multiple offices in one city or region | Combined peak fits in fewer buildings | Commute impact and team disruption |
Office consolidation is the biggest prize and the biggest risk. Merging two half-empty buildings into one well-run one can transform the cost base, but only if the combined midweek peak genuinely fits, with headroom for growth and for the days when everyone shows up. This is where a booking platform earns its keep after the decision too: capacity limits, waitlists and check-in data stop the consolidated office tipping into overcrowding, and give you early warning if demand starts to outgrow the space.
Subletting is the middle path when the lease will not flex. It only works if the underused space is separable, so use floor-level utilisation data to prove which floors can close cleanly. Renegotiation is the quiet win: walking into a lease event with a quarter of hard utilisation evidence changes the conversation with the landlord entirely.
A practical 90-day sequence
- Weeks 1 to 2: Build the full occupancy cost figure and calculate current cost per desk. Agree the target range with finance.
- Weeks 3 to 14: Capture desk and room utilisation across every site. Do not change policy mid-measurement; you want a clean baseline.
- Weeks 8 to 12: Attack the easy waste in parallel. Auto-release no-show bookings and convert chronically empty rooms to bookable desks or focus space.
- Week 13 onwards: Model the scenarios. What does the footprint look like at a managed 60 to 70% utilisation? Which floors close, which sites merge, which lease events are within reach?
- Then decide. Take the renegotiate, sublet or consolidate call to the board with the data attached, alongside the ongoing savings from better demand management.
The pattern across firms that get this right is consistent: measure first, manage the curve second, cut the footprint third. Occupancy cost is usually the second largest line after payroll, and it is the one where a quarter of honest data can move seven figures. Guesswork cannot compete with that.
Sources and further reading