HOME / CASES / RETAIL

10-time theft reduction in the checkout area and a revenue increase in a grocery store
Case on the implementation of the neural network Cyber Vision Control in a retail enterprise
This case will be of interest to those whose business is related to the retail industry. We are going to share a story of Cyber Vision Control implementation using the example of a grocery store. By the way, if the tasks of this case are close to you, please note that no more than 7 working days pass from signing the contract to the full launch of the program for this field of activity.
Contents list
A word on the customer
This customer is a small grocery store positioning itself with a motto "Always fresh." Store area: no more than 120 sq.m. Specialization: over-the-counter sale of meat and dairy products The departments are located at a distance of 9 meters from each other.

Please note that this was a pilot project for the store owner, on the basis of which it was planned to launch a whole chain of grocery stores.
There were three groups of problems to be solved
1. Effectiveness of the marketing department
Any marketing makes no sense if the introductory hypotheses turn out to be wrong. As the saying goes, if you expect results from marketers, then give them the facts. The facts include detailed and accurate information on the behavior of buyers and their preferences.

The degree of responsibility was also increased by the fact that a model of business solutions for all network objects was to be formed on the basis of one store. A pilot project always requires increased attention.

By the way, we think you are familiar with such a concept as visitor counters, and so, in this case, their use would not allow solving even part of the problem. One of the reasons is that a large number of random people passed through the store, including both non-target visitors and their own staff.
2. Cashbox discipline and dealing with theft
When we say "cashbox discipline", we mean the absence of theft of funds in the store. There are not so many situations in which financial cheating occur. Identifying them is a problem.

For example, a cashier opens a cash drawer, retrieves bills, but either the customer is absent or the "recount" button on the cash register is not pressed. A reason for reflection will be the fact that the cashier is already carrying out the tenth recount in the last couple of hours. Of particular interest is the cancel operation. Any cancellation is a reason to observe such an event. Cancellation in the absence of the buyer is also a reason for prompt action by the security service.

Or another method of theft, which is implemented in collusion with the buyer: 4-5 positions are broken through to the customer, and 6-7 positions (as a rule, the most expensive ones) go in the cart bypassing the cash register.
3. Labor discipline and compliance with regulations
We had to track the following events:
  • talking on the phone in the presence of the customer
  • lack of gloves and hats in the meat cutting area;
  • absence of sellers behind the counter;
  • absence of a manager when accepting goods.

A separate task was set to identify employees who were constantly hanging on their smartphones. We all know that time flies when one is on the phone. It seems you've just read a couple of posts on social media, and half an hour vanished into thin air. Now imagine if this is a girl at the counter or a cashier. A buyer comes up to her and waits for a young lady to stop messaging someone on her phone and to return to her duties. As a result there are not only queues, but also dissatisfied customers, who quite subtly feel the lack of respect for their time.

What neurorobots have learned
So, we trained our little helpers to perform the following operations:

  • counting store visitors and comparing with the number of receipts;

  • recognizing use of a smartphone/tablet by a cashier and generating a report on all such cases, including the time of using gadgets;

  • tracking those events at the cash register, which would record all types of theft and violations described abov

  • monitoring compliance with business processes in the goods acceptance area;

  • monitoring the fact of wearing hats and gloves in the meat cutting area.

CVC implementation results
the number of theft commited fell from 278 (first month) to 24 (third month)
After evaluating the data, the following changes have been implemented:
  • 1
    The system has learned to count visitors with an accuracy of 99.5%. This made it possible to form a cashier's work schedule, taking into account peak traffic hours, as well as to determine the conversion of buyers (with a small margin of error due to buyers who came together).
    • 2
      Customer loyalty has increased by improving the discipline of staff, including monitoring the use of gadgets.
    • 3
      The number of thefts at the checkout diminished by almost 10 times. In particular, we have practically eliminated theft in the process of fictitious cancellation of a receipt, return of goods, as well as fake cancellation of goods from a receipt. In the first month, 278 cases of such events were identified. In the second and third months there were 98 and 24 events, respectively.
    • 4
      Incidents of violating labor regulations in the meat cutting area has been reduced. Cases of not using hats/gloves dropped to almost zero (300 cases in the first month, 8 cases in the third month of system operation).
    • 5
      The cases of absence of the manager during the acceptance of the goods were practically reduced to zero (27 cases in the first month, 3 cases in the third month of the system operation).
    The cases of absence of the manager during the acceptance of the goods were practically reduced to zero (27 cases in the first month, 3 cases in the third month of the system operation).

    The accuracy of the program is 99.5%
    Thus, we organized tracking of almost 100% of possible violations at the enterprise, and also provided an opportunity to check the month of the enterprise's work in 3-5 hours of watching the prepared video clips for each detected case.
    Let's talk numbers.
    • Subscription fee: 15,000 rubles/month.
    • Equipment: 200,000 rubles.
    • Integration (implementation) cost: 80,000 rubles.
    In addition, we can safely state the following:
    • Reducing the number of lost customers;
    • Reducing the number of theft events by 10 times; ·
    • Growing revenue and profit of the enterprise

    Store/retail chain industry solution
    Cyber Vision Control
    CVC mitigates losses in any business, however small or big it is, whether it is a manufacturing enterprise
    or a service company.

    20 days
    Average pay-off period after CVC has been implemented
    >10 mln
    >100 mln
    Disciplinary violations detected
    Economic violations detected
    Read related
    Industry Solutions | Cases | Blog | About | Contact Us

    +7 (800) 301-39-70 contact@cvc.ai

    3, ul. Mira, ROOM 1H, St. Petersburg, 197101, Russia

    POWER AICONTROL© LLC
    OGRN (company number) 1197847127125

    Privacy policy