Know what your footfall is actually worth.
Your till tells you what sold. It cannot tell you how many people walked past the display and didn't stop, which entrance they came through, or whether last weekend's promotion pulled anyone in at all. Verkada's cameras answer those questions from hardware you were probably buying for security anyway — and DyCom deploys, tunes and supports it across your stores.
Most retailers run on two numbers: sales and staff cost. Everything between them is guesswork — how many people came in, what they walked past, where they queued, and which of your stores is genuinely busier rather than just bigger. Retail analytics closes that gap by treating the camera as a sensor: the same device that records an incident also counts the door, maps the floor and reports the hour.
The practical appeal is that it is one system. Counting, heatmaps, alerting and the footage itself live in a single cloud console, so a small team can run thirty stores without a VPN to each of them or a recorder in every back room.
What a store learns
All of it from the cameras, in one consoleTraffic counting
Digital line-crossing counts people through each entrance — and vehicles into the car park, which for a destination or precinct store is the number that actually predicts the day.
Occupancy Trends
How busy the store is by hour, so you can roster to the rush instead of to the timetable. The gap between when you think the peak is and when it really is tends to surprise people.
People Heatmaps
Customer movement plotted over your floorplan, showing which displays pull people in and which are dead space. Useful before a refit, and useful for proving to a supplier that their end-cap worked.
Conversion rate
Footfall exported by API and set against point-of-sale data, so you get the ratio that matters: of everyone who came in, how many bought. See the next section — this is the one worth doing properly.
Queue and service
Where people bunch up and for how long, so a second register opens because the floor data said so rather than because somebody noticed.
Store against store
The same metrics across every site in one dashboard, by day, week, month or a custom range — so a multi-site comparison is a filter rather than a spreadsheet exercise.
Footfall against the till
This is where the system either earns its keep or becomes a dashboard nobody opens. Counting people is easy; the number only becomes a business case when it sits next to revenue.
What it answers
- Of everyone who walked in, what share actually bought — your conversion rate, measured rather than assumed
- Whether a quiet day was fewer people or the same people buying less, which are completely different problems with completely different fixes
- Whether a promotion, a window change or a price move shifted traffic, conversion, or neither
- What a lease or a site is worth on traffic, not on a landlord's estimate
How it connects
- Verkada exposes the counts over an API, so the data is not trapped in the vendor's console
- From there it joins your POS or BI tool — the conversion figure belongs next to sales, where someone will actually look at it
- We do the join and the reporting, and agree up front which decision each number is meant to change
A counting system with no transaction data behind it tells you how busy you were. It never tells you what you missed.
Loss prevention, on the same cameras
Shrink and incident responseRetail theft and aggression toward staff have both risen, and the usual answer — more footage nobody has time to watch — does not help. What changes the outcome is being able to find the right thirty seconds quickly, and knowing when a known problem has walked back in.
AI search by description
Search footage by what someone was wearing or carrying — a jacket colour, a mask, a distinctive bag — instead of scrubbing a timeline. This is the feature that turns an afternoon of review into a few minutes.
Person of Interest alerts
Flag an individual so the system notifies the store when they next walk in, rather than after the fact. Retailers typically point this at repeat offenders in high-shrink sites. It is an identity feature — see the privacy note below.
Vehicle Search
Find a vehicle across the car park footage, which matters for organised theft where the vehicle is the thread that links incidents.
Incident Management
Package an incident with its footage and share it — including straight to police — without burning a disc or emailing a 4GB file. Faster handover tends to mean the matter actually progresses.
Remote oversight
Area and regional managers can look in on any store from the Command mobile app, so an incident does not wait for someone to drive there.
Retention that suits evidence
Hybrid storage — up to 365 days on the camera itself plus unlimited cloud archiving — so footage is still there when a claim or a prosecution surfaces months later.
Recognising the customers you want to see
VIP and priority arrivalsThe same alerting mechanism works in the other direction. Verkada can notify a manager by SMS, email or push when a priority customer arrives, matched on a photo, a licence plate or a written description. For high-value retail, a dealership or a members' club, the value is simply that the right person greets them by name instead of finding out they came in yesterday.
Worth being clear about what this is. VIP recognition and Person of Interest alerts are the same underlying capability pointed at different lists, and both of them identify individuals. That is a genuine step beyond anonymous counting, and it deserves a deliberate decision rather than a tick in a feature list.
The privacy conversation, stated plainly
Our people counting page makes a point of counting people and not identities, and that remains the right answer for most venues. This page is different, and we are not going to blur the two.
What that means in practice
- Traffic counts, occupancy and heatmaps are aggregate — no individual is identified to produce them
- VIP and Person of Interest alerting is not. It works by recognising a specific person, and that is the point of it
- Those are separate features you choose to enable. You can run the whole analytics side with none of the recognition turned on
How we scope it
- Identifying customers engages the Privacy Act and the Australian Privacy Principles — facial recognition in Australian retail has drawn real regulator attention, so this is not a settled area
- If you want it, it needs a written policy, proper signage and notice, a defined retention period and a named owner — not just a switch
- We will help you scope it, and we will tell you plainly if we think a site does not need it
Turn on what the business actually needs, tell people what is running, and be able to explain it. That is the whole test.
Why Verkada for multi-site retail
Where it isn't the right answer
We would rather say so nowBetter together with WiFi. Pair retail analytics with Ruckus WiFi presence analytics and you get the full picture — exact counts at the door, plus where people go and how long they stay once they are inside.
It is already running in Australian retail
The vendor's published reference, not oursCity Beach — an Australian lifestyle and fashion retailer trading since 1985, with more than 67 locations — runs Verkada across its stores, using Command, the mobile app, AI search, Occupancy Trends, People Heatmaps and Incident Management. The number worth quoting is the investigation time: around an hour down to about ten minutes, because staff can search by description instead of scrubbing footage. They are also testing Person of Interest alerts in sites with higher shrink.
"The ability to search for a person by clothing colour or send footage directly to the police made Verkada a standout solution."
Ben Nau, Head of Retail, City Beach — quoted in Verkada's published customer story. This is Verkada's reference, not a DyCom deployment; we include it because it is an Australian retailer at real scale rather than an overseas example.
Start with one store.
Tell us what you are trying to find out — conversion, staffing, shrink, layout — and we will scope the cameras, stand up the dashboard, join the counts to your POS and show you the numbers on a real trading week. Prove it on one site before you roll it out.
Verkada, Command, Occupancy Trends, People Heatmaps and Person of Interest are trademarks of Verkada Inc. Capabilities described are the vendor's; DyCom Technology is an integrator, not the manufacturer. Customer references are Verkada's published case studies.