This site stores cookies on your device according to our
Privacy policy
Tevian video analytics for performance control, loss prevention, and customer service in retail stores
Tevian technologies help retailers control key store performance indicators — from customer traffic and checkout load to shelf monitoring, customer demographics, and promotion effectiveness. The modules detect fraud, losses, and service issues, automatically send alerts, and provide access to real-time analytics. Integration with self-checkout systems and control systems is supported.
Marketing Analytics
Footfall analysis, customer demographic profile and traffic analytics
Tevian video analytics modules provide information on the number and types of visitors (new, regular, unique), return rate, as well as data on these counts by demographic characteristics and "typical customer profile", about areas of highest and lowest visitor interest, typical customer routes.
What does the system provide:
- Visitor counting with demographics
- The system determines the exact number of new, regular and unique visitors, excluding employees from the count.
- Repeat visits are recorded for a given period, return rate is calculated, which allows to assess customer loyalty and effectiveness of promotional campaigns.
- Metrics are supplemented with demographic segmentation (gender, age, ethnic group), available in convenient dashboards.
- Based on the data obtained, it is possible to evaluate the effectiveness of marketing campaigns, changes in the loyalty program, adjust the assortment and commercial offer to the typical customer profile of a specific store.
- Traffic analytics
- Heat map construction helps identify high and low activity zones of customers, assess display attractiveness and promo stand effectiveness.
- Multi-camera tracking shows complete movement routes from entrance to checkout, identifies blind spots and helps optimize store layout, improve promotional product placement and increase overall conversion.
Queue Control
Queue monitoring to improve service quality in stores
The system tracks checkout zone load in real time, records visitor crowds and allows quick response to queue increases. Analyzing video from cameras, the module automatically counts the number of people in queue, determines average waiting time and identifies control threshold exceedances. All data is stored and available as analytical reports for each checkout or store as a whole.
What does the system provide:
- Real-time cashier zone monitoring
- The system counts queue length and determines average waiting time for each checkout or self-service checkout.
- When set thresholds are exceeded, notifications are automatically sent to responsible employees.
- This allows to quickly open additional checkouts and avoid customer dissatisfaction.
- Cashier efficiency analytics
- Collected data is stored in the system and available as reports: you can analyze load by time of day, day of week or specific store in the chain.
- This helps identify "bottlenecks" during peak hours and optimize staff schedule.
- Customer experience improvement
- Reducing checkout waiting time directly affects customer loyalty level and perception of service quality.
Staff Monitoring
Monitoring staff presence and store operating hours
Tevian video analytics allows to control employee working hours, recording entry, exit and presence at workplace. Modules automatically determine store opening and closing time, help track lateness, early departures, breaks and overtime. All events are displayed in convenient reports and charts.
What does the system provide:
- Employee time tracking
- Presence and absence intervals are determined based on video, which allows to assess personnel efficiency.
- Archive video access
- Management can review recordings to check reasons for absence or confirm data.
- Discipline control
- Reports record entry, exit, lateness, early departures and overtime, display actual working time.
- Actual store opening and closing
- Determined by first and last person, allows to assess compliance with regulations.
- Centralized analytics
- Unified reports for all stores provide a complete picture of employee and store working hours.
Shelf Monitoring
Shelf fullness and product display quality control to increase profit
Tevian shelf analytics modules monitor shelf availability, detect out-of-stock situations, and verify price tag accuracy to ensure timely replenishment and correct merchandising.
What does the system provide:
- Shelf product availability control
- The system analyzes video streams at set intervals, determines shelf voids and calculates their fill percentage. When fill level drops below set threshold, notification is automatically sent to responsible employees. This allows to timely replenish products, reduce revenue losses and increase customer satisfaction.
- Price tag currency verification
- The system automatically identifies outdated or incorrect price tags. This helps control how quickly staff updates information on shelves according to trade matrix. Analytics can be used by administration or quality control department.
- Flexible data sources
- Incoming data can come from various devices: mobile phones, robotic platforms (e.g., vacuum robot) or stationary cameras. The system adapts to available infrastructure.
- Variative recognition scenarios, various algorithms are supported:
- Search for price tags missing in product matrix.
- Verification of price tag compliance with product category.
- SKU recognition and precise price verification against reference.
- Centralized analytics
- Unified reports for all stores provide a complete picture of employee and store working hours.
- Retailer receives an automated control tool for shelf fullness and price tag currency. This allows:
- Remotely evaluate personnel work.
- Exclude payment discrepancies.
- Increase customer trust and loyalty.
Antifraud
Loss prevention at checkouts and self-service checkouts, shoplifter identification
Tevian video analytics detects suspicious activities in checkout and self-checkout areas, including unscanned items, item cancellations, fraud during weighing, and bypassing checkout points. By correlating customer actions with POS system data, the technology identifies violations and sends alerts to security teams. The modules also recognize faces of potential shoplifters in video streams and compare them against an incident database. The solution automatically records and archives incidents, generates video reports, and allows quick review of fragments containing violations. Intelligent algorithms help prevent losses before they occur, while flexible configuration makes the module suitable for a wide range of retail scenarios.
What does the system provide:
- Suspicious action control at checkout and self-service checkout
- Events are recorded when product was handed to customer but not scanned in receipt, or item cancellation occurred. The system identifies such discrepancies using neural networks and saves them to video archive for post-analysis. Instant security service notification is possible.
- Events are recorded when product was handed to customer but not scanned in receipt, or item cancellation occurred. The system identifies such discrepancies using neural networks and saves them to video archive for post-analysis. Instant security service notification is possible.
- Flexible configuration for business tasks
- Administrator can set a list of fraud indicators: receipt cancellation, item cancellation, missing scanned product, etc. Custom integration with POS and self-service checkout systems is supported.
- Archive with quick violation viewing
- All suspicious actions are saved as video materials with filtering and jump to specific moment capability. This speeds up incident processing and helps take quick action.
- Self-service checkout and scale fraud
- Bypassing self-service checkout without scanning.
- Product substitution during weighing (e.g., "banana" instead of "mango").
- Incorrect selection in scale interface.
- Technology recognizes product by image and signals discrepancy.
- Shoplifter face recognition
- The system compares faces on video with known violator database and sends warning on match. Violators can be added manually. On repeat visit, the system instantly recognizes visitor.
- Important: blacklist management scenario, e.g., shoplifters, from legal perspective can be used in limited cases, e.g., when integrated with Safe City systems for public security.
Product Recognition on Scales
Faster service and fraud prevention during product weighing
Tevian’s neural network algorithms automatically identify the type of product on weighing scales using camera images. This reduces service time at checkouts and in the sales area, eliminates errors when selecting products, and prevents fraud at self-checkout. The system adapts to the store’s assortment and can be continuously trained on new SKUs.
What does the system provide:
- Reduced weighing and service time
- There is no need to manually search for products in the scale interface — the algorithms automatically recognize the product and suggest the correct option. This reduces the workload on cashiers and speeds up self-service operations.
- Fraud prevention at self-checkout
- The system detects cases where customers intentionally select a cheaper product (for example, choosing “banana” instead of “mango”). This reduces losses and ensures fair self-service transactions.
- System adaptability and learning
- The algorithm can be trained on new products: after several manual selections of a new item, the system will recognize it automatically, regardless of packaging type (plastic wrap, net, bag).
- Reduction of product misclassification
- Accidental errors by cashiers or customers when selecting products are eliminated, for example when weighing similar fruits, vegetables, or confectionery items.
- Improved customer loyalty
- Fast and accurate weighing without errors reduces customer frustration and speeds up the shopping process, especially during peak hours.
