Major Grocery Retailer Saves Time & Cost Using AI Powered Tools

Background

A major grocery retailer in the APAC region, with over $3 billion in annual revenue and more than 10,000 employees, embarked on an SAP implementation across its operations. The retailer's Point of Sales (POS) system is critical for ensuring seamless day-to-day transactions, managing approximately $10 million in daily revenue. However, during the SAP rollout, the organization faced significant challenges in testing the downstream impacts on its POS system.

Testing complex permutations of items, discounts, and promotions proved to be an overwhelming task. The POS system’s inability to interface directly with traditional testing tools further exacerbated the issue, making automated testing nearly impossible. As a result, testing cycles were delayed, manual testing efforts grew, and the risk of errors in customer-facing transactions increased—posing a threat to both operational efficiency and customer satisfaction.

Tricentis Vision AI

To address these challenges, TTC Global introduced Tricentis Vision AI, an advanced solution that uses computer vision techniques powered by AI frameworks like YOLO (You Only Look Once). YOLO is a real-time object detection system that processes entire images simultaneously, enabling rapid and accurate recognition of visual elements within the user interface (UI).

Vision AI uses this capability to identify and interact with UI components such as buttons, menus, and text fields in a way that simulates human testers. This allows for automated interactions with visually complex systems like the POS, even when direct system access is unavailable. By using this AI-driven framework, Vision AI adapts quickly to changes in the UI and can execute tests across multiple environments without significant reconfiguration. This makes Vision AI ideal for automating tests on systems that are otherwise inaccessible to traditional tools.

TTC Global’s Solution

TTC Global deployed a tailored solution that combined Tricentis Vision AI with additional automation techniques to streamline both regression and volume testing for the POS system. This approach significantly reduced the time and effort required for testing while improving accuracy.

Regression Testing

TTC Global automated over 100 test cases using Tricentis Vision AI, focusing on complex scenarios such as multiple discount types, including club deals, and overlapping promotions. Leveraging a data-driven approach, they automated a variety of item and discount combinations, ensuring consistent coverage of multiple pricing scenarios.

A critical aspect of the solution was the integration of Test Data Service (TDS), which dynamically updated data sets for testing. This allowed the automated tests to reflect real-world business scenarios in real time. The regression suite was executed regularly to validate the system as new changes were introduced, reducing manual test execution time from 3-4 working days, down to 11 hours, or a decrease of approximately 75% in total elapsed time.

Call out featuring the key metrics here? To be formatted in an eye-catching way

• 100 test cases automated

• Approximately reduced testing time by 75%

Volume Testing

Beyond regression testing, TTC Global also automated volume testing using API-driven approaches. The POS system's "change lane" functionality was accessed through APIs, while department-specific working hours were also managed via APIs. The team performed data querying through APIs to retrieve the necessary information, which was then set up in SAP to execute comprehensive volume tests.

The automated volume testing suite ran through 2,000 articles, a task that would have been impractical to perform manually. The automated process completed testing in just 6.5 hours, a significant efficiency gain for the retailer.

Summary

By implementing Tricentis Vision AI and advanced automated software testing techniques, TTC Global helped the retailer fully automate both regression and volume testing, achieving faster and more accurate testing cycles. The use of AI-driven computer vision allowed for automation on complex systems where traditional tools failed. With over 100 test cases automated and volume tests running at a fraction of the time required for manual testing, the solution delivered substantial time savings and enhanced system stability. By implementing this approach, the retailer was able to allocate resources to other strategic initiatives, all while ensuring the accuracy and quality of their critical POS system.