Cameras in the store count incoming footfall, queue length and whether a staff member is in the service area. Once the queue passes the threshold, the manager gets a message and opens a second till. We do not recognise faces: we count people, we do not identify them.
At six in the evening there are five people at the till and the second one is closed. Two of them leave their baskets and walk out.
The sales assistant went to the stockroom, and a customer spends ten minutes looking for someone to ask.
Nobody knows how many people came in and how many bought: there is no door counter, and receipts tell a different story.
Shift schedules are copied from last month, even though the peaks have long since moved.
What changes after launch
01The store manager knows about a queue while customers are still in it
02Shift schedules follow real peaks, not guesswork
03You see how many people came in and how many reached the till
How the system works
01
We mark store zones
Entrance, tills, key service areas. Together with the store manager we decide what to count in the pilot.
02
Footfall counting
How many people came in by hour and day. We count people but do not identify them: no biometrics are collected.
03
Queue watch
We set a threshold — for example, more than 4 people at the till. Once it is passed, the manager gets a message.
04
Unstaffed areas
If nobody from the staff is in a service area for too long, it goes into the shift log. The rules and access to this data are agreed with you in advance.
05
Summary for planning
Peaks by hour and day, plus a store heatmap if needed — for shift schedules and store layout.
One task, three different projects. Pick the one that looks like you and press “I want this” — your request arrives tagged, and we send an estimate in 1–2 days.
Small business
One store or branch where evening queues have become the top complaint in reviews.
What we build
1 store, 1–2 zones
Incoming footfall counting
Queue detection and manager alerts
Summary by hour and day
Integrations
Telegram, in-store cameras
Where it runs
Edge unit in the store, or cloud if your security policy allows it
Timeline
Guideline: 3–6 weeks for the pilot
Mid-size business
common start
A chain of several stores where each manager reports in their own way and there is nothing to compare stores by.
What we build
Several stores in one system
Footfall and queues compared across stores
Footfall linked to receipts from your POS or ERP (1C)
Dashboard for store and regional managers
Support contract: new stores and zones
Integrations
ERP (1C), POS system, Telegram
Where it runs
Cloud or your own server, depending on security requirements
Timeline
Guideline: 2+ months, starting with one store
Enterprise
common start
A regional or national chain with its own VMS, a security team and a rule that video must not leave the perimeter.
What we build
Integration with your VMS across all stores
One chain-wide dashboard broken down by region
Roles and permissions: a store manager sees their own store
Event and action log for audits
Service under an SLA
Integrations
VMS, ERP (1C), POS and BI systems
Where it runs
Your servers or a closed perimeter: no face recognition, in line with Russian data law (152-FZ)
Timeline
Guideline: 3+ months, starting with a pilot in one store
Pilot from RUB 500,000. Beyond that we price by task and hours.
Questions
Is this staff surveillance?
The scenario is about service and zone utilisation. Goals and data access are agreed with you before the start. We do not do staff monitoring as a separate task without a dedicated specification.
Do you recognise customers by their faces?
No. The system counts people and measures queues but does not identify anyone. No biometrics are collected, and if video must not leave the store, processing happens on site.
Do we need new cameras?
The ones already above the entrance and tills often work. During the assessment we review your footage: if the angle does not allow counting, we will tell you where to put a camera.