A store has a door. Count everyone through it, subtract the exits, and you know how many people are inside. A plaza has no door. A park has a dozen ways in and long stretches of open edge where the boundary between inside and outside is a matter of where you decide to draw it. That single difference changes everything about how you measure a public space, and it is the reason a method built for a shop entrance does not simply transfer to an open square.
Cities increasingly want the answer anyway. When a square is redesigned, a park re-landscaped, or a seating scheme installed, someone has to say whether it worked, and "it feels busier" is not an answer a capital budget respects. Occupancy data turns a redesign from a matter of taste into a matter of evidence: when is the space busiest, which corners never get used, does a new bench change how long people stay, and should the cleaning crew come at the hour the space is actually full rather than the hour the timetable says.
This post is about measuring open public spaces specifically: plazas, squares, parks, and greens, not a shop and not a pedestrianised shopping street. It covers the core distinction between counting people who cross a line and measuring how many are present in an area, the dwell signal that says whether a space works, what the data actually drives, and how you count an open outdoor space without a camera. A pedestrianised retail street is a related but different case, and destination tourism flow is different again; this post stays on non-retail open space.
How do you measure occupancy in a plaza or park?
An open public space has no single doorway, so occupancy is measured differently from a store. One approach counts people crossing the entry lines around a defined area and keeps a running balance of who is inside; another samples how many people are present in zones of the space over time to build an occupancy and dwell picture. The data answers practical questions: when is a plaza busiest, which corners of a park go unused, does a new seating or planting scheme change how long people stay, and when should maintenance and cleaning run to match real use rather than a fixed timetable.
Those two approaches, counting the boundary and sampling the interior, are not interchangeable, and choosing the wrong one for the question is the most common mistake in open-space measurement. The next section pulls them apart.
Line-crossing counts vs area occupancy: two different questions
The two questions sound similar and are not. Line-crossing counting measures flow: how many people passed a point in a given direction over a period. Area occupancy measures presence: how many people are inside a defined area at a given moment. A busy thoroughfare across a square can have very high line-crossing counts and very low occupancy, because everyone is passing through and nobody stays. A small sunny corner can have modest flow and high occupancy, because the few who arrive settle in for an hour.
You can derive occupancy from line-crossing if, and only if, you can count every way in and out of a defined area and keep an accurate running balance of entries minus exits. That works well for a plaza with a countable set of access points, where sensors on each entry maintain a live tally of who is inside. It works badly for a sprawling park with an open perimeter, where there is no complete set of lines to count and the balance drifts as small errors at each crossing accumulate over a day. Counting groups correctly at those entry lines is its own technical problem, because people arrive in twos and threes and a naive counter miscounts them; the accuracy factors are covered in counting groups accurately.
For open-perimeter spaces, sampling presence within zones of the area is the more honest method: rather than trying to balance an incomplete set of boundaries, you measure how many people are present in each zone and read occupancy and dwell directly. The practical rule is simple. If the space has a countable boundary, line-crossing with a running balance gives you both flow and occupancy. If it does not, measure presence in zones and do not pretend an incomplete boundary count is a true occupancy figure.
Dwell and lingering: the metric that tells you a space works
Occupancy tells you how many people are in a space. Dwell tells you how long they stay, and for a public space that is the metric that actually says whether the design works. A plaza people cross is infrastructure; a plaza people linger in is a place. The whole point of investing in seating, shade, planting, and programming is to move a space from the first category toward the second, and dwell is the number that registers whether the move happened.
Lingering is also where the value of a public space shows up in ways a count cannot. A square where people stay supports the cafes and kiosks around it, hosts the events a city wants to programme, and generates the sense of safety that comes from a space being occupied rather than empty. Reading dwell across the day shows the rhythm of a space: the lunchtime fill and empty of an office-district plaza, the long weekend afternoons of a park, the dead hours a redesign might target. This is close cousin to how a destination reads visitor dwell across a whole place, the subject of city tourism flow analytics, though here the frame is a single space.
The signal a city is usually chasing is a rise in dwell without a fall in turnover: people staying longer while the space still serves everyone who wants it, rather than a few people monopolising it. Occupancy and dwell read together catch that. Rising dwell with steady occupancy is a space working better; rising occupancy with falling dwell is a space busier but less pleasant to stay in, which is often the first sign a design is being overwhelmed.
What the data drives: maintenance, safety, programming, design
Occupancy and dwell data is only worth collecting if it changes a decision, and in an open public space it changes several concrete ones.
- Maintenance timing. Cleaning, waste collection, and servicing scheduled to the hours a space is actually busy, and done in the genuine lulls, instead of a fixed timetable that sends a crew through at peak or misses the mess entirely.
- Safety and crowding. Knowing the busiest hours and the corners that fill lets a city plan lighting, sightlines, and stewarding around real occupancy rather than assumptions, and spot when a normally quiet space is filling unusually.
- Programming. Placing markets, performances, or pop-ups in the hours and zones that have capacity to fill, and evidencing turnout afterward, so the programming calendar is built on where and when people actually are.
- Design evaluation. Reading occupancy and dwell before and after a change to a seating layout, planting, or surface, so a redesign is judged on whether people used the space more and stayed longer, not on how it looks in a render.
Each of these is a case where a fixed assumption gets replaced by a measured pattern, and each one compounds: the same continuous data that schedules the cleaning also evaluates the redesign and sites next season's market.
Counting an open outdoor space without cameras
Measuring a plaza or a park means measuring people in a space that is public by definition, where the objection to a camera is loudest and most legitimate. People sit in a square precisely because it is an open civic space, not a monitored one, and a city that bolts image-capturing cameras to it to count heads invites exactly the backlash that gets a scheme cancelled. Camera-free measurement is what makes counting an open public space acceptable, and because these spaces are used into the evening and after dark, the method also has to work without daylight.
Ariadne measures this with Hybrid Fusion, its patented camera-free method. Time-of-Flight depth sensing counts every visitor at the entrances, capturing geometry rather than images, while patented phone signal sensing follows movement through the interior, detecting the signals a phone emits even in airplane mode, and tracks that movement to about one-metre precision. The sensor streams both feeds to Ariadne, where Hybrid Fusion combines them into one trajectory per visit and computes counts, dwell, and paths. The streams carry no identifier: no MAC address, no device ID, no biometric data, and no camera is involved. Identifiers are stored only when a visitor explicitly opts in, which keeps the method GDPR-friendly and outside biometric territory.
For an open space the outdoor and after-dark performance is not a footnote, it is the requirement, because Time-of-Flight measures distance rather than relying on light and keeps counting through the evening a park or plaza is often at its liveliest. The details of counting reliably outdoors and in low light are covered in counting outdoors and after dark. The privacy posture and the outdoor reliability are the two reasons the method fits public space where a camera does not.
From measurement to a design or funding case
The reason to measure an open space is usually to make a case with the result: to justify a redesign, to win the capital for a park refit, or to show a funder that money already spent moved the needle. Occupancy and dwell data does that in a way a photograph of a busy afternoon cannot, because it is continuous, comparable over time, and honest about the hours and corners that do not work as well as the ones that do.
The strongest cases pair a before-and-after occupancy read with a dwell read, on the same measurement points, over comparable seasons. A redesign that lifts both occupancy and dwell against its own baseline is defensible evidence the space works harder now; one that lifts occupancy but drops dwell tells a more careful story worth knowing before the next scheme repeats the mistake. Presenting the trend rather than a single strong day is what separates evidence from anecdote, and it is what a funder deciding on the next space is asking for. For how continuous, camera-free measurement supports a portfolio of public spaces across a city, see how people counting feeds smart-city analytics.
FAQ
How do you measure occupancy in a plaza or park?
By either counting people across the entry lines of a defined area and keeping a running balance of who is inside, or by sampling how many people are present in zones of the space over time. A plaza with a countable set of access points suits the line-crossing approach; a park with an open perimeter is better measured by presence within zones.
What is the difference between line-crossing counts and area occupancy?
Line-crossing measures flow, how many people passed a point over a period, while area occupancy measures presence, how many are inside an area at a given moment. A busy thoroughfare can show high flow and low occupancy because people pass through; a sunny corner can show low flow and high occupancy because people stay.
Why does dwell time matter for a public space?
Because a space people cross is infrastructure and a space people linger in is a place. Dwell registers whether seating, shade, planting, or programming moved a space from one to the other, and it explains the value a raw count misses: the trade, the events, and the sense of safety that come from a space being used rather than passed through.
Can you measure park and plaza occupancy without cameras?
Yes, and in public space it is usually necessary. Ariadne uses Hybrid Fusion, a camera-free method that captures geometry rather than images and no MAC address by default, which answers the privacy objection an open civic space raises, and because it measures distance rather than light it keeps counting outdoors and after dark.
Do I need cameras to count a public space?
No. Ariadne counts with Hybrid Fusion: Time-of-Flight depth sensing plus patented phone signal sensing, never cameras. Time-of-Flight captures geometry rather than images, and signal sensing captures no MAC address by default, so the measurement involves no video, no faces, and no biometric data.
