Occupancy counting in buildings has become a central lever for building energy performance, occupant safety, and the technical management of commercial, industrial, and public sites. Driven by the proliferation of IoT sensors, building occupancy management is taking on a new dimension: it is no longer just about counting people, but about cross-referencing this data with energy consumption, air quality, or the BMS (Building Management System) to turn every square meter into actionable data. Between regulatory obligations (the Tertiary Decree, the BACS Decree, GDPR) and the operational expectations of facility managers, occupancy counting in buildings is becoming a strategic issue for local authorities, landlords, property companies, and real estate departments.
Table of contents
Measuring building occupancy addresses several challenges that, before the arrival of IoT, were rarely combined within a single tool.
The first is energy efficiency. A building that is air-conditioned, heated, or lit according to a fixed schedule consumes energy even when the premises are empty. Occupancy counting makes it possible to adjust heating, ventilation, and lighting to the actual presence of occupants, making it one of the fastest levers for meeting the building energy performance trajectories imposed by the Tertiary Decree.
The second challenge is people's safety. In buildings open to the public (known as 'ERP' under French regulations), counting occupants is a regulatory requirement under the French Building and Housing Code and fire safety regulations: knowing the number of occupants in real time is essential to complying with evacuation thresholds and managing emergencies.
The third challenge is organizational. Optimizing building occupancy also means better sizing workspaces and adapting cleaning, maintenance, or fire safety measures to actual usage, while documenting how spaces are used for future fit-out or flex-office projects.
Finally, occupancy counting feeds into the BMS (Building Management System) and the CTM (Centralized Technical Management), providing usage data that, when cross-referenced with energy data, enables predictive rather than reactive management.
Counting building occupants relies on two inseparable building blocks: the sensors that capture the flow, and the IoT platform that turns this raw data into an actionable indicator.
Several families of IoT sensors can be used to measure building occupancy, each with a different level of accuracy, cost, and regulatory constraint:
A sensor on its own only produces a signal. It is the hypervision platform that gives meaning to occupancy counting in buildings, by aggregating data from multiple heterogeneous sensors and systems (BMS, CTM, access control, air quality) within a single dashboard. This software layer makes it possible to visualize occupancy in real time, configure threshold alerts, cross-reference occupancy data with building energy performance data, and keep a usable history for regulatory reporting, including the annual declaration on the OPERAT platform required under the Tertiary Decree.
The choice of counting technology depends on the building, the level of accuracy required, the budget, and data protection constraints. The table below summarizes the main trade-offs.
| Technology | Counting accuracy | GDPR / privacy impact | Deployment cost | Typical use case |
| Infrared / presence sensor | Medium (flow, no identity) | Low, no personal data | Low | Meeting rooms, restrooms, corridors |
| Anonymized counting camera | High | Moderate, processing must be regulated (CNIL) | Medium | Lobbies, building entrances, buildings open to the public |
| CO2 / air quality sensor | Indicative (density, not exact count) | None | Low | Enclosed rooms, open-plan spaces |
| Wi-Fi / BLE | Medium to high depending on device density | Moderate, anonymization required | Low to medium | Large floors, shopping centers |
| Badge / access control | Very high but limited to badge holders | High, personal data | Medium to high | Offices, secure sites |
| Sensor coupled with BMS/CTM | High, integrated into technical management | Low | Medium | Automated heating, ventilation, lighting |
Beyond the raw technology itself, the decisive criterion is often interoperability: the sensor's ability to feed its data into a hypervisor capable of cross-referencing it with the BMS, the CTM, and the building's other business systems, rather than remaining confined to a proprietary application silo.
Kuzzle offers a modular architecture built around three complementary building blocks: Kuzzle Data, an open-source foundation for collecting and securing heterogeneous data; Kuzzle IoT, dedicated to connecting, decoding, and supervising connected devices; and Kuzzle Hypervision, the application layer that centralizes business data — including occupancy counting data — into unified dashboards.
In the smart building field, this architecture makes it possible to bring BMS, CTM, air quality, energy, and space occupancy data together into a single point, in order to automatically trigger optimization scenarios: reducing heating or air conditioning in unoccupied areas, adjusting lighting based on actual presence, or issuing predictive maintenance alerts before equipment failure. The platform is available in a low-code configuration, allowing operations teams to create new occupancy indicators themselves without relying on custom development, and with sovereign hosting, in a French cloud or on-premise.
Several deployments illustrate this approach in practice. In Noisy-le-Grand, Kuzzle IoT was used to connect public buildings and monitor air quality and energy consumption. In the Alès urban area, a territorial hypervisor centralizes air quality and energy consumption monitoring across several sites. SIEA relies on a hypervision platform to supervise air quality and electricity data.
In the hospital sector, a university hospital uses Kuzzle's real-time geolocation to optimize the management of its connected medical equipment and improve patient care — a real-time monitoring approach that is directly transferable to occupancy counting in corridors and waiting areas.
The Côte-d'Or department has deployed a control and automation demonstrator across 150 public buildings.
For a facility operator, the benefit of this approach is twofold: on one hand, having a single hypervisor avoids multiplying supervision interfaces by sensor type or building; on the other hand, it strengthens regulatory compliance, since centralizing occupancy and energy data makes it easier to produce the indicators required by the Tertiary Decree and the BACS Decree.
Occupancy counting and building occupancy management fall within a precise legal framework that combines energy regulations, fire safety, and personal data protection.
The Tertiary Decree, stemming from decree no. 2019-771 of 23 July 2019 and codified in articles L174-1 to L174-3 and R174-22 to R174-32 of the French Building and Housing Code, requires commercial buildings larger than 1,000 m² to reduce final energy consumption by 40% by 2030, 50% by 2040, and 60% by 2050, compared with a reference year between 2010 and 2019.
Consumption data must be declared every year, before 30 September, on the OPERAT platform managed by Ademe (the French Environment and Energy Management Agency), under penalty of administrative sanctions of up to €1,500 for an individual and €7,500 for a legal entity, as well as a public "name and shame" procedure.
The BACS Decree, codified in articles R175-1 to R175-9 of the French Building and Housing Code (notably article R175-2 for the applicability threshold), requires the installation of a Building Automation and Control System — generally a BMS — in commercial buildings equipped with heating or air conditioning systems with a capacity above 70 kW. The deadline is set at 1 January 2025 for installations above 290 kW. For those between 70 and 290 kW, the deadline, initially set at 1 January 2027, was postponed to 1 January 2030 by decree no. 2025-1343 of 26 December 2025, in order to align with the European timetable.
This postponement does not, however, call into question the energy reduction targets of the Tertiary Decree, nor the value of getting ahead of compliance, since the current equipment rate across the building stock remains very low. The text specifies that a BMS alone is not enough: the system must enable consumption to be monitored by functional zone and deviations to be detected, which directly aligns with the need for occupancy counting and monitoring.
The RE2020 environmental regulation, which came into force through the order of 4 August 2021, complements this framework for new buildings by setting energy and environmental performance requirements from the design stage onward, encouraging the native integration of sensors and management systems.
At the European level, Directive 2024/1275 of 24 April 2024 on the energy performance of buildings (EPBD) reinforces these obligations and pushes member states to roll out automation and control systems across the commercial sector more broadly, with a gradual transposition expected into French law.
Regarding fire safety in buildings open to the public, the safety regulations against fire and panic risks, set out in articles R123-1 et seq. of the French Building and Housing Code, require that the number of people present be known at all times in order to comply with the evacuation thresholds specific to each category of public-facing building.
Finally, whenever occupancy counting relies on sensors capable of processing personal data (camera, Wi-Fi, BLE, badge), the General Data Protection Regulation (EU Regulation 2016/679) applies. The CNIL (the French Data Protection Authority) distinguishes three regimes: near-immediate anonymization of data (within a few minutes), based on legitimate interest, with no consent required; pseudonymization with deletion or anonymization at the end of the day, accompanied by clear information and a right to object; or, failing that, the collection of free, informed, and specific consent from the individuals concerned. In all cases, only aggregated and anonymous statistics should be retained, sensitive areas (staff-only rooms, for example) must be excluded from the measurement scope, and any attempt at re-identification is prohibited.
Yes, provided the CNIL's recommendations are followed: rapid data anonymization, informing the people present, excluding sensitive areas, and avoiding any attempt at re-identification. Most occupancy counting solutions for buildings rely on aggregated processing that retains no individual data beyond a few minutes or the current day.
The BACS Decree requires the installation of a building automation and control system in sites where heating or air conditioning capacity exceeds 70 kW, with consumption monitored by zone. It does not explicitly require an occupancy sensor, but occupancy counting is, in practice, the most effective complement for achieving the management objectives set out in the regulation.
A BMS (Building Management System) controls the technical equipment of a single site: heating, ventilation, lighting, access. A CTM (Centralized Technical Management) extends this logic to several buildings or sites, centralizing supervision within a single hypervisor, which is particularly useful for local authorities or property companies managing an extensive real estate portfolio.
For a first deployment, infrared or presence sensors coupled with a hypervisor are often the best entry point: moderate deployment cost, no personal data, and simple integration with an existing BMS. Moving to more precise technologies, such as anonymized counting cameras, generally comes later, for areas with higher footfall.
A hypervisor centralizes occupancy counting, BMS, CTM, energy, and air quality data within a single interface, whereas this data is often still scattered across multiple proprietary tools today. This centralization makes it possible to trigger automatic actions (adjusting heating, ventilation, lighting) and to directly produce the indicators required for regulatory reporting, particularly under the Tertiary Decree.