Phone signal tells you a suburb. Bookings tell you an intention. Neither tells you whether a room is full at 11:40 on a Thursday.
Phone-carrier signal tells you a suburb. Card spend tells you a transaction, hours later. Bookings tell you an intention, which is not the same as a person in a room. Social posts tell you about the small fraction of a night somebody chose to perform.
None of them answer the only question that matters at 11:40 on a Thursday: is this room full? So the measurement layer is ours, end to end — the sensor on the wall, the hub in the building, the backhaul out of it, and the models at the other end.
A battery sensor on the wall of a venue. It counts presence, not people — no camera, no audio, nothing that leaves the room as an identity. Months of life on a single cell.
Collects every leaf in range, timestamps it against a common clock, and holds the night locally if the connection drops. A venue never loses a Saturday to a dead uplink.
Cat-1 bis out of the building on its own connection. A venue never has to put us on their Wi-Fi, their card network, or their IT team's roadmap.
Occupancy against a venue's own baseline, movement between precincts, and the forecasts that let an operator see tonight before tonight happens.
This is a hardware constraint, not a setting. There is no mode in which a leaf records an image, a voice or a name, because it has no camera, no microphone and no concept of a person.
We map where the crowd actually stands — bar, floor, smokers' area, queue — and place leaves to match. Most venues need a handful.
Leaves are battery powered and adhesive-mounted. No cabling, no ceiling work, no electrician, no closing early.
One hub per venue on a standard socket, with its own cellular connection. Nothing touches your network.
The first fortnight establishes your ordinary week. After that the dashboard is reading tonight against your own history, not a city average.