Product · n7kel Flow
n7kel Flow — visitor flow analytics
How many people came, where they wait, how long they spend at a service window, an ATM or a consultation, and where they go next. n7kel Flow turns video from the cameras you already have into a picture of visitor flow for banks, stores, hospitals and offices. For customers who have provided their face themselves and given consent, individual visit analytics is available.
- 110:02Entrance
- 210:03–10:11Waiting area · 8 min
- 310:11–10:19Desk 2 · 8 min
- 410:20–10:23ATMs · 3 min
- 510:24Exit
All other visitors: anonymous counts only
Capabilities
What n7kel Flow shows
Footfall
Visitor numbers by hour, day and site, entries and exits, peak hours.
Queues
Queue length in each zone and a notification when a queue exceeds the threshold, so another window or checkout can be opened in time.
Waiting time
How long visitors wait before being served and how this changes during the day.
Time in zones
How much time people spend in the waiting area, at a service window, at an ATM, in the consultation area, on the sales floor.
Routes
How people move around the site: from which zone to which, where they stop, which zones stay empty.
Heat maps
Zone occupancy on the site plan — for decisions on staffing, service windows and equipment placement.
Banks
Individual visit analytics — for customers who have given consent
A bank branch serves a large flow of customers every day. Customers who have provided their face to the bank themselves — for example, when signing up for services — and have consented to analytics are recognized at the entrance. For them, n7kel Flow shows how the visit went:
- how long the customer spent in the waiting area;
- how long service at the window or a consultation took;
- whether the customer used an ATM or a self-service terminal;
- in what order the customer visited the branch zones.
All other visitors are counted anonymously — only as headcount, waiting time and zone occupancy, without establishing identity. If the wanted-list check module is enabled at the site, it works separately — only under an agreement with the authorized bodies.
How it works
How n7kel Flow is built
Cameras and site plan
The cameras you already have are used. Zones are marked on the plan: entrance, waiting, service windows, ATMs, consultations, sales floor.
Anonymous counting
n7kel AI counts people and time in zones without establishing identity. This is the base mode for all visitors.
Consent module
Individual analytics for customers who have provided their face and given consent is enabled separately and only by the client’s decision.
Analytics and notifications
A dashboard by site and network, queue notifications, report export, data passed to your systems.
Scenarios
Typical scenarios
| Site | Question | What n7kel Flow provides |
|---|---|---|
| Bank branch | Are there enough service windows at peak hours? | Queues and waiting times by hour, long-queue notification |
| Bank branch | How does a regular customer’s visit go? | Time in zones and order of visits — only for customers who have given consent |
| Store | Which areas of the floor work and which stay empty? | Heat map, routes, time in zones |
| Store | When should additional checkouts be opened? | Real-time checkout queues, peak hours |
| Hospital, outpatient clinic | Where do patients wait longest? | Waiting time at reception and outside doctors’ rooms; with n7kel Access, check-in by appointment |
| Business center, office | Load on entrance areas and elevator lobbies | Footfall by hour and entrance |
Privacy and the law
Recognition only with consent
- visitor recognition is off by default: the base mode of n7kel Flow is anonymous counting;
- individual analytics is enabled only by the client’s decision, as a separate module, and only for customers who have provided their face themselves and given consent;
- a customer can withdraw consent — the reference image is deleted, and from then on the customer is counted anonymously;
- biometrics is a special category of personal data: the client is the data controller, and reference images are stored within the client’s environment;
- any restrictions needed — zones, retention periods, who has access — are set at the client’s request;
- when the module is enabled, the face of a person entering is compared with the reference images of customers who have given consent; no reference images are created for other visitors; the processing scheme at the entrance — including how visitors without consent are kept anonymous — is agreed by the client’s lawyer before the module is enabled;
- all data stays with the client, is confidential and is not shared with third parties, except where expressly provided for by the agreement and the law; the final legal assessment is made by the client’s lawyer.
Integration
Connecting to your systems
- the bank’s queue management system and CRM — to match a visit with the service provided;
- the store’s POS system — for visitor-to-purchase conversion;
- the hospital’s appointment system — together with n7kel Access;
- BI and reporting — REST API, webhooks, CSV and XLSX exports;
- n7kel Atlas — for analyzing flows across the entire network of sites.
Pilot
How we begin
Site survey
The site plan, cameras, zones and the questions that need answering.
Legal framework
For individual analytics: the procedure for obtaining consent and an assessment by the client’s lawyer.
Pilot
2–4 weeks at 1–2 sites, anonymous mode from day one.
Report
Figures against the agreed criteria and recommendations on service windows, staffing and zones.
Questions
Does n7kel Flow recognize every visitor?
No. The base mode is anonymous counting. Only customers who have provided their face themselves and given consent are recognized, and only if the client has enabled this module. To do this, a face at the entrance is compared with the reference images of customers who have given consent; no reference images are created for other visitors, and the processing scheme is agreed by the client’s lawyer before the module is enabled.
Will our cameras work?
As a rule, yes. n7kel AI is trained and tested on ordinary 2D images, so the cameras you already have are suitable. 3D cameras are supported too, if they are already on site or the task requires them, for example for accurate counting at the entrance.
Can we see the whole branch network?
Yes. Data from all sites is combined into network-wide analytics, and with n7kel Atlas connected it is linked to events from other systems.
Let’s look at your visitor flow
Tell us about your sites and what you want to know about the flow, and we will propose zones, metrics and a pilot format.
The interfaces in the illustrations are schematic mockups with sample data.