"Crossroads" together with GoodsForecast increases the turnover of stores

2022: Implementation of GoodsForecast.OSA

The introduction of the GoodsForecast.OSA solution, which controls the availability of goods on shelves, helped the Perekrestok retail chain to increase retail turnover by 1.4% in supermarkets selected for participation in the pilot project. This was announced by GoodsForecast on November 1, 2022.

By the start of the project, Perekrestka had already used a product availability control solution that generates signals for the best-selling goods and promos. But it did not allow timely elimination of display defects and ensure the availability of the necessary goods on the shelves.

Illustration: my-perekrestok.ru

To solve its problems, Crossroads chose GoodsForecast. Before the start of cooperation, the vendor had developed practices and implemented similar projects in other networks, confirming the effectiveness of the tools used. The industry expertise possessed by the GoodsForecast team also became a significant argument.

The project included the introduction of the Alerts module of the GoodsForecast.OSA solution to generate signals to stores across a wide range of online products. The system receives check data during the day at a specified frequency, allocates periods without sales and determines for each article how much this corresponds to the usual demand for a specific product in a given store. Then the system generates a signal that comes in the form of a task to check the goods in the hall, and sends it to the mobile device of store employees assigned to the corresponding department. The employee of the department checks whether everything is in order with the availability, price tag, availability and quality of the goods, and in case of a problem takes the necessary measures.

Illustration: kiosksolutions.ru

The Crossroads partnership with GoodsForecast started in August 2021. At the first stage, experts set up integration data processes, including sales and balances on an hourly basis. Further, on the basis of collected and marked up data, models of availability assessment were built algorithm , signals for warning shops and ranking of issued signals were improved in order to maximize potential additional revenue. This system was integrated with the Perekrestka microservice, which provides signal exchange and reverse communication with retail outlets. To assess the economic effect of the system, a specialized division of X5 Group ad hoc developed an A/B testing design lasting 2 months. A pilot group was formed, which included about 132 stores with a wide geography. A control group was selected for the selected outlets to evaluate the results. The target metric of the experiment was the positive dynamics of retail turnover of stores. X5 Group specialists calculated that for statistical significance it should be at least 0.8%.

According to the test results, according to the X5 Group ad hoc assessment, the model showed a statistically significant result: the point estimate of the RTO increase was 1.4%, and the upper limit of the confidence interval was 2.4%. Also, the specialized division of X5 Group concluded that with a full-scale deployment, the same results of growth in turnover will be achieved, provided that the quality of processing of incoming signals by store employees is maintained.

From October 1, 2022, the signal generation module was launched in all Perekrestok supermarkets, and the project has already been developed. The X5 Group and GoodsForecast team are working on restoring demand based on signals, followed by reverse integration in order to increase the forecast accuracy in the Perekrestok network auto-order system.

The increase in turnover by more than 1% gives a significant financial effect on a network scale. However, in addition to the financial component, there are other impressive results. When arriving at supermarkets, the buyer expects to see the necessary goods on the shelves in sufficient quantities and with a good shelf life. If that doesn't happen, everyone loses. The client is left without purchases for which he came, and the network does not receive revenue. The ability to ensure the availability of goods on the shelves allows not only to increase sales, but also to meet the expectations of buyers, to maintain their loyalty. Therefore, the project with GoodsForecast has become unambiguously effective, commented Kirill Kovalenko, Owner of the product Mobile workplace of the Perekrestok retail network.

In business, in practice, most decisions are still made on the basis of expertise, not on the basis of data analysis. The partner project was carried out according to all the rules, which is the great merit of the Perekrestok network team and their very progressive approach. The company received a result confirmed by business and mathematics, told Lev Kiselev, Head of Retail Business analysts - GoodsForecast.