Turn foot traffic into numbers you can act on.
You can't manage what you can't measure. People counting gives retail, hospitality, councils and venues an accurate, anonymous count of who comes through — so you can staff to demand, prove conversion, and stop guessing about the busiest hour of your week.
Every business has a rough feel for how busy it gets. Very few can put a number on it — or tell you what that number did to sales, staffing cost or the queue at the counter. People counting closes that gap with sensors at the door and the zones that matter, feeding a dashboard that answers the questions a till roll can't: how many walked in but didn't buy, when the rush actually hits, and whether that promotion pulled anyone through the door.
One thing that makes us unusual: almost every people-counting company in Victoria resells a single platform, so the recommendation is settled before anyone looks at your building. We carry three — and we will tell you when none of them is the right answer for your door.
What it measures
The metrics behind the floorFootfall
Accurate in-and-out counts by entrance, hour, day and season — the true top of your funnel, not a till count that misses everyone who left empty-handed.
Live occupancy
How many people are inside right now, per space — for capacity limits, safety, and knowing when to open a second register.
Conversion rate
Footfall against transactions, so you can see what share of visitors actually buy — and whether a change to the floor or the roster moved it.
Dwell time
How long people stay, and in which zones — the difference between a browse and a buy, and a read on where attention goes.
Queue & wait
Queue length and wait time at the counter, so you can trigger more staff before customers walk out rather than after.
Zones & heatmaps
Which parts of the venue pull traffic and which are dead — evidence for layout, merchandising and where the promo really belongs.
Direction & tailgating
Counts that hold up under real conditions — directional in/out logic, and alerts when someone moves the wrong way or two people pass as one.
Vehicles & car parks
The same counting logic applied to entries, exits and bays — useful wherever the car park is the real front door.
The platforms we deploy
Three, because one doesn't fit every siteA council library, a national retail chain and a stadium concourse have different accuracy, privacy and budget constraints, and the honest answer changes with them. Here is what each is genuinely good at.
Venue analytics at scale
- Australian platform — airports, centres, stadiums, councils
- LiDAR-based counting: no camera, no image captured
- WiFi presence analytics read alongside the door counts
- ISO 27001 certified, 200+ third-party integrations
Counting at the edge
- AXIS People Counter — bidirectional counting on the camera itself
- AXIS Occupancy Estimator — live occupancy and average visit time
- Object Analytics counts people and vehicles
- Analysis on-device: no footage leaves the building
One pane of glass
- Occupancy Trends across every camera and site
- Crowd Notifications when a space exceeds a threshold
- Heatmaps and floor-plan people maps for space planning
- Analytics sharing a platform with access control
The accuracy question, answered straight
You will be told, correctly, that camera-based counting is less accurate than a purpose-built overhead counter. Here is the honest version, because you will hear it in any competitive quote and it deserves a real answer rather than a dodge.
What the numbers say
- A dedicated 3D stereo or time-of-flight counter, correctly mounted at 2.5–4.5m, sits around 98–99.5%
- Repurposed general security CCTV counting is commonly cited around 80%
- Legacy infrared beam sensors fall below 85% and should be replaced, not extended
What it means for you
- If the number underpins a rent review or funding submission, use a dedicated counter — we will specify one
- If you want trend and zone insight from cameras you were buying anyway, camera analytics is the economical answer
- A purpose-built application on a correctly positioned camera is not the same as pointing generic CCTV at a door — but it is still not a 3D counter
Anyone who tells you their camera matches a dedicated overhead counter at the doorway is selling, not measuring.
Those are published industry ranges, not our own testing. Whatever we install, we validate it against a manual hand count before you rely on it — that step matters more than the spec sheet.
How an install actually goes
Start with the question, not the sensor
What decision is the number going to change — a roster, a lease negotiation, a layout, a capacity limit? That determines where sensors go and how accurate they need to be. Sites that skip this step buy hardware and get dashboards nobody opens.
Scope the room
Ceiling height, entrance width, glazing, sunlight, traffic volume, and whether staff cross the same doorway forty times a shift. We survey the space and pick the sensor and mounting that will hold accuracy under those conditions.
Install, calibrate, validate
Mount, aim, configure the in/out lines and zones, then validate against a hand count before anyone trusts a report. Calibration is the difference between a counting system and an expensive guess.
Integrate and report
Pull the counts alongside POS, roster or booking data so the numbers sit next to the thing they explain. Dashboards for the floor, scheduled reports for the board, alerts for the moments that need someone now.
Counting people, not identities
The fastest way to lose a council tender or a centre contract is to be vague about privacy. So we are not: these systems count bodies crossing a line. They do not identify anyone, and they are not a surveillance product dressed up as analytics.
What is captured
- Anonymous counts — a person crossed a line, in this direction, at this time
- Aggregate occupancy, dwell and zone density
- Directional logic that filters staff re-entries and double counts
What is not
- No facial recognition and no biometric matching
- No personally identifiable information stored against a count
- Overhead 3D and LiDAR sensors have no recognisable image to store in the first place
Where a deployment uses cameras rather than dedicated counters, we will tell you plainly what the camera can do beyond counting, and scope it so the system does only what you have told your visitors it does. Australian Privacy Principles apply to most of these deployments, and signage obligations are part of the design, not an afterthought.
What it's worth
Who uses it
Better together with WiFi. Pair people counting with Ruckus WiFi presence analytics and you get the full visitor picture — exact counts at the door, plus where people go and how long they stay once they're inside.
Questions we get asked
How accurate is it, really?
It depends far more on the install than the brand. A correctly mounted overhead 3D or LiDAR counter over a normal doorway is very accurate; the same sensor aimed badly at a five-metre glazed atrium is not. That is why we survey first and validate against a hand count before handing over — and why we will tell you when a site needs a different sensor than the one you were quoted elsewhere.
Do we need cameras?
No. Overhead 3D and LiDAR counters are not cameras and store no recognisable image — often the right answer for libraries, councils and anywhere the privacy conversation matters more than the extra features. Camera-based counting earns its place when you want the security footage anyway, or need zone heatmaps and queue analysis a door counter can't give you.
Can it use the cameras we already have?
Sometimes. Axis analytics run on many existing Axis cameras, so if you already have them in the right positions the software may be most of the job. Positions chosen for security are frequently wrong for counting, though — a camera angled to catch faces at the door is rarely angled to count reliably through it. We will tell you honestly which of your existing cameras can be reused.
Does it need to talk to our POS?
Not to work, but conversion rate is where counting starts paying for itself, and that needs transactions alongside traffic. Most POS and booking systems can export or expose what we need; we handle that integration as part of the build rather than leaving you with two dashboards to eyeball side by side.
Why DyCom rather than a people-counting specialist?
Two reasons. First, most specialists in this market are tied to a single platform, so the recommendation is decided before they see your site — we carry three and pick to the room. Second, a counter is a PoE network device with a dashboard attached: it needs an address, a switch port, patching, monitoring, and someone to notice when it drops off. A specialist hands that back to your IT provider. We are your IT provider.
Who supports it afterwards?
The same team that runs managed IT, networks and WiFi for our clients — 25+ years, Melbourne-based, part of the DyCom Group. One number to call, whether the problem is the sensor, the switch it hangs off, or the report that stopped arriving.
Put a number on it.
Tell us the space and what you're trying to learn — conversion, capacity, staffing, queue times — and we'll scope the sensors, stand up the dashboard, and quote it up front. Start with one door and grow from there.