AI Case Study

Coca-Cola Hellenic covers 3.6X more stores by automating in-store inventory audits using image recognition

Coca-Cola Hellenic uses image recognition technology, deep reporting and real-time analytics to keep track of SKUs. Trax's platform helps accelerate retail execution by automating store audits. Coca-Cola Hellenic has reduced audit times by upto 9 times. The reduction in costs has enabled them to triple their sales execution coverage and reduce out of stock.

Industry

Consumer Goods And Services

Food Beverage And Drugs

Project Overview

"Integrating AI applications into field operations enables end-to-end efficiency improvements that can unlock revenue opportunities and significantly increase share of shelf.

Before sales reps even reach the stores on their route, AI applications can determine the optimal order for store visits based on the best time to visit each one, available personnel, in-store promotions and specific store needs.

Within the store, digital image recognition can be used to streamline store audits by capturing information about SKUs, tasks that previously depended on time-consuming and error-prone manual entry. Less time spent on data gathering and product placement and distribution leaves more time for what reps are actually meant to do – active selling.

The gathered information can be quickly analyzed to evaluate out-of-stock items, facings, prices and share of shelf, and even find buried patterns that can point to sensible next steps"

Reported Results

Company claims:

* Reduced out-of-stock incidents by 63%
* Increased execution scores by more than 10%
* Achieved 97% SKU recognition accuracy

Technology

Function

Marketing

Merchandising

Background

"Headquartered in Zug, Switzerland, Coca-Cola Hellenic Bottling Company, is one of the world’s largest bottlers of drinks from the Coca-Cola Company. They operate 56 plants and 264 distribution centres in 28 countries, selling more than 2 billion unit cases (50 billion servings) every year."

Benefits

Data

"Retailers use fixed cameras to capture shelf images every few minutes. Those images are uploaded where Trax machine-learning algorithms and engines identify every product, price, and promotion."