Shelf Recognition AI changes the way of managing exposition at DANONE Poland


Solutions that use artificial intelligence to manage the availability of products on store shelves are often referred to as the sixth sense of sales reps. This is great that this type of systems is also popular in our country. DANONE group of companies has just reached for eLeader Shelf Recognition AI, the Polish patent. Together, we make sure to recognize the needs of the brand consumers faster and respond to them more accurately.

Managing the availability of products on the shelf

eLeader Shelf Recognition AI is one of the fastest tools on the market. By combining the recognition technology with the reporting panel, we provide users with a functional system for analysis, optimizing store displays and verifying the availability of goods.

Our solution improves and optimizes the work of sales representatives in the field. The system analyses the parameters, responsible for the correct display of goods on a store shelf, while real-time data helps representatives make right decisions. Thanks to this, they are able to respond to the consumers’ needs ‘right here, right now’, which has a positive effect on the sales volume. The data, provided through the Shelf Recognition AI neural networks, also supports the management of the sales chain and allows to efficiently react to the challenges of the constantly changing market. At Danone Poland, we support the daily work of 170 sales representatives throughout the country. Every month, the algorithm of our system processes over 120 thousand photos, which allows to recognize 460 products, including both Danone’s own products and those of their competitors. Identified goods are sorted into 14 separate product categories.

 For quite a while, the market, especially when it comes to FMCG, has been looking for tools that would  provide precise information, related to the presence of products on store shelves. The availability of such data is of great importance, because over 75% of purchasing decisions are made in the store. This means that if the product is not available or its display does not meet the consumer’s expectations, it can lead to the abandonment of purchase or the selection of competitors’ goods. Therefore, ensuring that high standards of product exposition are implemented has a huge impact on sales results, which is recognised by companies that already use advanced data analysis solutions – says Rafał Melanowicz, Head of eLeader AI Department

eleader-shelf-recognition-ai-zmienia-oblicze-handlu-w-danone

Artificial intelligence supports salespeople

A sales rep uses our solution when there is a need to examine the display of products on the shelf. By taking photos, the representative sends research materials to the neural network, on the basis of which the algorithm provides reliable information about the share of shelf parameter. In addition, it verifies presence, evaluates the number of facings, indicates out-of-stocks and provides a full picture of all products in a given category, including information about direct market competitors. The entire analysis is completed just within a few minutes, and its result allows the representative to adjust the exposition during the same visit. Therefore, this solution is much faster, compared to similar activities, carried out in a traditional manual way.

New technology makes the work easier

By implementing artificial intelligence, DANONE has significantly improved the current model of collecting and analysing the data. According to Łukasz Pawłowski, Senior Sales Analysis and Systems Development Team Manager at DANONE:

– When implementing eLeader Shelf Recognition AI, we primarily wanted to reduce the time of manual reporting, performed by our sales representatives. We sought to introduce an innovative tool that would provide non-stop access to fast, automatic and complete information about the situation on the market (in the stores), and indicate key areas of operation to our representatives during their visit to the point of sale. eLeader application enables us to increase the number of surveyed indicators and, what is important, it does not require the extension of salespeople’s working hours. The implemented tool also contributes to increased availability of our products on the shelves and improved quality of their display. We focus on innovation in what we do, which is why we are one of the first companies in our category to implement eLeader Shelf Recognition AI.

Other benefits are mentioned by Dariusz Radzimirski, Senior Business Solutions Manager at DANONE – The solution is fully integrated with SFA and the data seamlessly flows to other IT systems. This makes managing the entire system easier and also allows to scale the solution up to other countries where DANONE has operations.

Prior to the implementation of Shelf Recognition AI, the company had been already using the SFA system, developed by eLeader. The applied perfect store model introduces key performance indicators (KPIs) and calculates points for meeting the standards. The solution allows to evaluate a store, a representative or a visit in just a few minutes.

Information about the model, implemented on the Polish market, was well received by other international branches of DANONE, which follow the development of solutions, based on the image recognition technology, with interest.

The case study and the full history of eLeader Shelf Recognition AI implementation at DANONE can be found here.