Most workplace floor plans are built around a headcount that no longer exists. A department signed a lease for 200 desks in 2019, and today 70 of them sit cold every morning while three meeting rooms near the coffee point get fought over daily. The lease still reflects the old number. The cleaning contract, the HVAC schedule and the security patrols all reflect it too. Workplace occupancy analytics exists to close that gap between the building you pay for and the building people actually use.
Occupancy analytics tracks how many people are in a space, where they cluster, and when. That sounds simple until you try to act on it. A raw people-count tells you a room was busy; good analytics tells you it was busy between 10am and noon on Tuesdays and Thursdays, empty every Friday, and consistently over its stated capacity by 30%. That pattern is what changes a budget decision.
There is a difference between presence detection and counting. A motion sensor tells you a room is occupied or not. A counting sensor tells you how many. For workplace decisions you almost always need the second kind, because a six-person room used by one person is a utilisation problem a motion sensor will never surface.
If your data is wrong, every decision built on it is wrong, and confidently so. This is where vendors get vague. Vemco, which has been doing this since 2005, works to a contractual minimum of 96% accuracy, typically reaching 98–99% when conditions allow. Those conditions matter: lighting, ceiling height, entrance layout and how people move through a doorway all affect the result. A busy double-door entrance where people walk two abreast is harder to count than a single controlled turnstile.
Ask any vendor who promises a flat 99% under every condition to show you the test setup. Real accuracy is a range, not a slogan. One practical detail worth insisting on: staff exclusion. In a university library or a public service centre, if the counter includes the reception team and the cleaners, your peak-load figures drift upward and you over-provision. Sensors that separate staff from visitors give you numbers you can plan against.
The clearest financial case is always-on systems running against empty rooms. Lighting, heating, cooling and ventilation running full schedule in zones nobody uses is money leaving the building every hour. When occupancy data connects to your HVAC and building management system, those schedules follow real demand instead of a fixed timetable.
This is what VemFusion is built for — connecting live occupancy data to HVAC, BMS and security so a wing that empties out at 3pm stops being conditioned as if it were full. VemSpace handles the other half: the space utilisation and facility optimisation view that shows you which floors, rooms and desks earn their footprint and which do not.
The savings tend to fall into three buckets:
Occupancy analytics serves two clocks. The slow clock is the quarterly trend: how utilisation shifts across a semester, a seasonal peak, or a return-to-office push. That view drives lease and design decisions.
The fast clock is the live count, with an alert that fires the moment a space crosses a predefined limit. For a public building or a lecture hall with a hard capacity, that alert is a safety and compliance tool, not just a convenience. Facilities teams use it to redirect people before a space is dangerously full, and to prove afterward that limits were respected.
Here is something that catches teams off guard: the first two weeks of data usually look wrong, and they are not. Occupancy patterns are noisier than people expect. Meeting rooms are double-booked but half-attended. The desk your directors swear is always occupied turns out to be a hot desk used three days a week. Do not act on week one. Let a full cycle of behaviour settle — a working month for an office, a full term for a university — before you sign off on a reconfiguration. The teams that skip this step end up removing desks people needed and adding rooms that stay empty.
Also worth planning early: privacy positioning. Counting people is not the same as identifying them, and saying so clearly to staff and unions removes most of the resistance before it starts. Sensors that count anonymously and exclude staff are far easier to defend than any camera-based system tied to identity.
Start with one question your finance director already asks: are we paying for space and energy we do not use? Then instrument the areas where you suspect the answer is yes — the underused floor, the meeting rooms, the entrances feeding the busiest zones. Measure against real accuracy figures, connect the output to the systems that spend money, and let a month of data make the argument for you.
Occupancy analytics is not about surveillance or gadgets. It is about matching what a building costs to how a building is used, with numbers accurate enough to defend in a budget meeting.
If you want to see what your own floors are really doing — and where the always-on waste is hiding — talk to the Vemco team about a workplace occupancy analytics setup built around your building's layout and accuracy conditions.