Occupancy Sensors for the Office: A 2026 Buyer's Guide

Deciding where and how to place occupancy sensors for office space is now a core part of any real estate or facilities strategy. Hybrid work has made static assumptions about how many people are in the building on any given day unreliable, and lease costs are too high to keep paying for space nobody uses. This guide explains how the technology actually works, where each sensor type fits, and what to check before you sign a purchase order.

What occupancy sensors are (and are not)

An occupancy sensor detects whether a space is being used, and increasingly, by how many people. The simplest devices answer a binary question: occupied or vacant. More capable systems produce a live count, dwell time, and movement patterns that feed occupancy sensing analytics.

It helps to separate two things people often conflate. Presence detection tells you a desk or room is in use right now. Utilization measurement aggregates that signal over weeks to reveal how a floor, neighborhood, or building is really performing. A lighting-control PIR sensor does the former well and the latter poorly. When your goal is space decisions, you want a system built for continuous real-time occupancy monitoring, not just switching lights off.

Sensor types compared

No single sensing technology wins on every dimension. Accuracy, range, privacy, install effort, and cost all trade against each other. Here is how the main options behave in practice.

PIR / passive infrared

PIR sensors detect the heat and motion of a body moving through a detection cone. They are inexpensive, low-power, and mature.

• Pros: cheap, simple, long battery life, well understood.
• Cons: they detect motion, not presence. A person sitting still to read or take a call can register as absent, producing false vacancies. They also cannot count people.

Radar / mmWave

Millimeter-wave radar emits a low-power radio signal and reads the reflections. Because it senses micro-movements like breathing, it holds a room as occupied even when people are still.

• Pros: strong presence accuracy, can estimate people counts and zones, works in the dark, captures no imagery.
• Cons: higher unit cost than PIR, more careful placement and tuning required, higher power draw for some models.

Under-desk sensors

These small devices mount beneath a desktop and detect whether someone is seated at that specific workstation, usually via infrared or a short-range radar element.

• Pros: excellent for individual desk occupancy sensor data; unambiguous per-desk attribution.
• Cons: one device per desk means higher device counts and more batteries to manage at scale.

Thermal

Thermal sensors use a low-resolution infrared array to read heat signatures as blobs rather than recognizable images.

• Pros: good counting accuracy, privacy-preserving by design because no identifiable image is captured, works in darkness.
• Cons: more expensive, and heat sources like radiators or sunlight can introduce noise if placement is careless.

Camera-based computer vision

Instead of adding hardware, computer vision analyzes feeds from cameras you may already have installed and turns them into anonymized counts and heatmaps. Modern systems process frames on the edge and output numbers, not stored video.

• Pros: rich data (counts, flow, queue length, zone dwell) with no new sensor per space; reuses existing infrastructure.
• Cons: requires clear sightlines and adequate camera coverage, and demands a disciplined privacy posture that anonymizes at the source.

Wired vs wireless, and why the network matters

Sensing accuracy means little if the devices are painful to deploy or keep failing to report. This is where the network layer quietly decides the success of a rollout.

Wired sensors offer constant power and reliable connectivity, but running cable to every desk and doorway is expensive and disruptive in an occupied building. Wireless removes the cabling problem but introduces two constraints you must plan for: battery life and radio range.

Consumer wireless protocols such as Wi-Fi and Bluetooth were not designed for hundreds of low-traffic sensors reporting small packets. They drain batteries quickly and congest the same spectrum your staff rely on. This is why purpose-built platforms run on a private PULSE IoT LoRaWAN network. LoRaWAN is a low-power, long-range protocol: a single gateway can cover a large floor plate, walls and all, while devices sip power and last years on a battery rather than months.

• Battery life: replacing coin cells across a building is a real operational cost. Multi-year battery life is the difference between a system you install once and one that becomes a maintenance burden.
• Coverage: LoRaWAN's range means fewer gateways and simpler site surveys, especially in dense or hard-to-reach areas like stairwells and basements.
• Independence: a private network keeps sensor traffic off corporate Wi-Fi and under IT's control.

Desk vs room vs zone sensing

Match the granularity of sensing to the decision you are trying to make. Over-instrumenting is expensive; under-instrumenting leaves blind spots.

• Desk sensing: per-workstation signals power desk booking auto-release and precise ratios for shared-desk programs. Best served by under-desk sensors.
• Room sensing: a meeting room sensor confirms whether a booked room is genuinely in use and how many people are inside. mmWave and thermal both work well here.
• Zone sensing: for neighborhoods, cafeterias, and collaboration areas, you care about aggregate flow rather than individual seats. Radar zones and computer vision are the natural fit.

Privacy considerations

Occupancy data touches employees, so privacy is not an afterthought. The guiding principle is data minimization: collect the least identifiable signal that still answers your question.

PIR, radar, thermal, and under-desk sensors are inherently anonymous, they never capture an image. For camera-based approaches, insist that recognition and counting happen on the edge and that only aggregated, anonymized numbers leave the device, with no stored or streamed footage tied to individuals. Be transparent with staff about what is measured and why, honor regional rules across India, the US, and the GCC, and keep the stated purpose narrow: space optimization, not surveillance of individuals.

From signal to decision

Sensors are only useful if the data drives action. Two workflows deliver most of the value.

The first is desk booking auto-release. When a colleague books a desk or room but never shows, the seat sits reserved and unusable. Feed live presence into the booking system and a no-show reservation releases automatically after a set grace period, returning capacity to the pool without anyone chasing it.

The second is longer-horizon space planning. Weeks of space utilization sensors data reveal which neighborhoods run hot, which meeting rooms are chronically overbooked while others sit empty, and what your true peak occupancy is versus your seat count. That evidence turns lease renewals, floor consolidation, and layout changes into decisions grounded in occupancy analytics rather than anecdote.

Buyer's checklist

Use this list to compare vendors on the factors that actually determine outcomes.

• Sensing accuracy: does it detect stationary presence and produce reliable counts, not just motion?
• Network: is it a private, low-power LoRaWAN deployment with multi-year battery life and whole-floor coverage?
• Granularity: can it mix desk, room, and zone sensing on one platform?
• Privacy: is data anonymized at the source, with clear handling that meets your regional requirements?
• Integration: does it connect to desk and room booking to enable auto-release out of the box?
• Analytics: are utilization dashboards and exportable trends included, not a costly add-on?
• Scale and support: how simple is device management, firmware updates, and expansion across sites?

Conclusion

The right occupancy sensors for office space depend on the questions you need answered, the granularity that matches them, and a network built to run reliably for years without babysitting. Weigh sensor types on presence accuracy and privacy, favor low-power LoRaWAN for a maintainable wireless deployment, and make sure the data flows into booking and planning workflows where it creates value. If you want to see how integrated IoT and computer vision sensing come together on one platform, explore UrSpayce occupancy sensing.