Face Recognition
Cuts Retail Losses

Client review

«In a substantial number of cases, we were able to return the money for the unpaid goods upon the shoplifter’s reappearance in the stores of our network, and there was no need to contact the MOI. This is how, thanks to the use of the STOP Shoplifter system with facial recognition module from NtechLab, the Podruzhka chain recovered over 500 000 rubles in six months. In the future we plan to scale up the system’s use to most of the chain’s stores.»

Vitaly Sotnev
Chief of security at the Podruzhka retail chain

The project in numbers

Within 6 months of the solution’s operation, more than 160 shoplifters have been detained in the Podruzhka store chain.
50+ stores
equipped with face recognition system
160 shoplifters
caught in 6 months of the program’s operation
Over 500 000 rubles
in recovered damages

Partnership outcome

With the help of NtechLab’s facial recognition technology, BIT has created a market-ready product for the retail sector called «STOP-Shoplifter». The software is successfully used in the largest retail chains, including dozens of stores of the Podruzhka chain, providing protection of retail outlets from «forgetful» customers. During the first 6 months of operation in the chain’s stores it allowed to detain more than 160 shoplifters and save over 500 000 rubles.

Top 5 shoplifter-favorite goods at «Podruzhka» stores:

  • Nail design kits
  • Nail polish
  • Lipsticks
  • Body and hair oils
  • Eye shadow

The Podruzhka beauty store chain, comprised of 224 stores.

BIT, the developer of the STOP-Shoplifter solution for effective theft prevention with the help of facial recognition.


1. To prevent theft

2. To minimize losses caused by unscrupulous customers

3. To get a tool for pre-trial conflict resolution


Violators’ images taken during the theft or from the CCTV archive are uploaded to the system. The system’s cloud server is integrated with the FindFace facial recognition algorithm. When an unscrupulous customer returns to the store (according to statistics, this happens at least in 50% of cases), the software notifies the security personnel.