Most buildings are designed once, for a version of their occupants that never quite shows up. The floor plan is a hypothesis: this many desks, this many meeting rooms, a canteen sized for a lunch rush that turns out to peak twenty minutes earlier and half as hard as anyone drew. Human-centred design is the discipline of closing that gap, of shaping a space around how people actually use it rather than how a plan assumed they would.
This is a deep dive on human-centred design (spelled human-centered design in the US), why understanding space user patterns is the part everyone skips, and how you measure those patterns honestly without turning your building into a surveillance operation.
01 · WHAT HUMAN-CENTRED DESIGN ACTUALLY MEANSWhat human-centred design actually means
Human-centred design (HCD) is an approach that makes systems usable and useful by grounding them in an explicit understanding of the people who use them, their tasks, and their environment. That definition is not marketing language. It is close to the wording of ISO 9241-210, the international standard for human-centred design, which formalised decades of practice from usability and ergonomics research.
The idea has deep roots. Don Norman popularised user-centred design in The Design of Everyday Things, arguing that when people struggle with a door or an interface, the fault is the design, not the person. Design studios like IDEO turned the same instinct into a repeatable method: observe real people, define the actual problem, prototype, test, and iterate. The through line is simple. You do not design for the user you imagine. You design for the user you observe.
ISO 9241-210 sets out six principles for a human-centred approach:
- The design is based on an explicit understanding of users, tasks, and environments.
- Users are involved throughout design and development.
- The design is driven and refined by user-centred evaluation.
- The process is iterative.
- The design addresses the whole user experience.
- The design team includes multidisciplinary skills and perspectives.
02 · FROM PRODUCTS TO SPACES: WHY BUILDINGS NEED IT TOOFrom products to spaces: why buildings need it too
Human-centred design grew up in product and software teams, but a physical space is a designed system too, and a far more expensive one to get wrong. A floor plan encodes assumptions about behaviour: where people will gather, how they will move, when a zone fills and empties. The moment the building opens, those assumptions meet reality, and reality rarely reads the brief.
The built-environment world has a name for principle three applied to space: post-occupancy evaluation (POE). POE is the practice of studying a building after people move in, to connect how it performs with how it is actually experienced. Reviews of POE in workplaces and campuses consistently find the same thing, that design intent and real occupant behaviour diverge, and that the only way to know how is to measure the occupied building rather than trust the drawing. Occupancy data, as one practitioner guide puts it, moves space design from guesswork to evidence.
03 · THE GAP BETWEEN DESIGNED USE AND ACTUAL USEThe gap between designed use and actual use
The gap is not small, and it is not a rounding error. It is the single largest source of wasted floor area and wasted energy in commercial property.
Industry benchmarks put average global desk utilisation near 43%. Density's analysis of workplace data found that roughly a quarter of desks go completely unused on any given day, and that most occupied desks are used for under three hours. In hybrid offices, average occupancy commonly sits between 40 and 60 per cent, far below the peak the space was sized for. Gensler reports that a large majority of corporate real-estate leaders now rank space utilisation as their key metric for workplace decisions.
That is the human-centred design problem stated in floor area. A building sized for its busiest imaginable hour spends the other ninety per cent of the week conditioning, lighting, and cleaning space that nobody is in.
04 · UNDERSTANDING SPACE USER PATTERNSUnderstanding space user patterns
Averages hide the story. A floor at 45 per cent utilisation is not 45 per cent full everywhere, all day. It is packed in the collaboration zone at 11am, dead in the far corner by 3pm, and quietly ignored on Fridays. Human-centred design needs the shape of use, not a single headline number. In practice that means four patterns:
- Peak versus average. Sizing to the peak wastes money. Sizing to the average leaves people standing. You need both curves to make the trade honestly.
- Zone-level density. Which specific areas fill, which never do, and how that changes by hour and by day. This is where reconfiguration decisions actually live.
- Dwell time. How long people stay tells you whether a space is a thoroughfare, a focus zone, or a social anchor. The same square metres serve very different needs.
- Flow. How people move between zones exposes the desire lines a static plan cannot predict, the shortcut everyone takes and the meeting room nobody walks to.
Read those four together and a building stops being a floor plan and starts being a behaviour you can design for. That is the whole promise of human-centred design applied to space.
05 · HOW TO MEASURE SPACE USER PATTERNS WITHOUT SURVEILLANCEHow to measure space user patterns without surveillance
Here is the catch that stops most human-centred design programmes before they start. To understand how people use a space, you have to observe them, and the obvious ways to observe people at scale are exactly the ways that erode their trust.
The usual options each carry a cost:
- Surveys and manual audits capture how people feel, which matters, but they are periodic, subjective, and expensive to run often.
- Badge and booking data show who tapped in or reserved a room, not who actually used it. Ghost bookings and tailgating make the numbers soft.
- Wi-Fi and Bluetooth counting infer presence from devices, which breaks under address randomisation and, when identifiers are re-linked, quietly turns headcount into tracking. We wrote about why device-based occupancy tracking struggles on both accuracy and privacy.
- Cameras with cloud video are accurate but store and transmit footage of identifiable people, which is a compliance and reputation liability few facilities teams want to own.
There is a fifth option, and it is the one Occivar was built around: on-device edge AI. An occupancy sensor processes each frame on the device, produces an anonymous count, and discards the frame in milliseconds. No video is stored, no video is transmitted, and no faces are recognised. You get the optical accuracy of a camera with the privacy profile of a presence detector, because the raw image never leaves the room. That is what our on-device edge AI does, and it is the difference between measuring a space and surveilling it.
06 · TURNING PATTERNS INTO HUMAN-CENTRED DECISIONSTurning patterns into human-centred decisions
Measurement is not the goal. Better decisions are. Once you can see the real pattern of use, the human-centred moves become obvious and, more importantly, defensible in front of a finance team.
Give people the spaces they actually need. A 2024 Leesman study found that in offices offering only basic desks, 41 per cent of employees said they could work productively, while in offices with a considered mix of focus pods, meeting areas, and quiet nooks that figure rose to 85 per cent. The lesson is not more space. It is the right space, informed by how people work.
Right-size the systems that follow occupancy. Heating and cooling is the clearest example. When you drive plant from real occupancy instead of a fixed schedule, you stop conditioning empty floors. Our occupancy-based HVAC control use case and the worked example from a real building show the mechanism and the savings in full.
Reconfigure the dead zones. The corner that never fills is not a failure of the people. It is a signal. Live occupancy analytics turn that signal into a floor-by-floor map of what to keep, merge, or repurpose, and audit-grade reporting turns it into the evidence a NABERS submission or a board paper needs.
Every kind of space has its own rhythm. A university library empties and fills on a timetable a retail floor would not recognise, and a hospital ward reads differently again. The same anonymous, human-centred method adapts across all of them, which is why it runs across the industries we work in.
07 · HUMAN-CENTRED DESIGN IS A LOOP, NOT A LAUNCHHuman-centred design is a loop, not a launch
Return to principle four from the standard: the process is iterative. This is the part that separates human-centred design from a one-off fit-out. A space is never finished, because the people in it keep changing how they work. Hybrid schedules shift, teams grow, a new tool changes where collaboration happens. A plan that was human-centred in 2024 can quietly stop fitting by 2026.
That is why continuous, anonymous occupancy data matters more than a single post-occupancy study. Measure, change, and measure again. The building becomes a feedback loop instead of a fixed bet, and each cycle gets you closer to a space that fits the humans in it. For a portfolio of buildings, small improvements compound into serious money and serious carbon.
08 · SOURCES AND FURTHER READINGSources and further reading
- ISO 9241-210:2019, the international standard defining human-centred design and its six principles.
- IDEO, on human-centred design as a practical method of observe, prototype, and iterate.
- XY Sense Workplace Utilization Index, for the global average desk utilisation benchmark.
- Density space-utilisation benchmarks, on unused and under-used desks.
- Gensler Global Workplace Survey, on utilisation as the leading workplace metric.
- Post-occupancy evaluation research, on the gap between design intent and occupied behaviour.
Human-centred design is not a mood board. It is the discipline of building around evidence of how people actually use a space, and evidence is exactly what most buildings never collect. If you want to see the real pattern of use in your own building, anonymously and without a camera storing a single frame, book a demo and we will walk it with you.