Case Study
Accelerating Fashion Retail Testing For Leading Fashion Brands
FireFlink helped a fashion brand automate tough testing challenges, making script development faster and test management easier. Learn how FireFlink transformed their testing process. Read on.
Transformation Summary
FireFlink transformed the testing process for a leading fashion retailer by enabling faster script development, automating critical customer-facing features, and centralizing test management across web and mobile platforms.
Problem Statement
One of the most loved fashion brands with 324+ stores spread across 200+ towns and cities in the country, offering a versatile collection & retail apparel with over 100 licensed and international brands, including exclusive in-house brands; The organization is often referred to as the fashion powerhouse with an elegant bouquet of leading fashion brands and retail formats. Present across 31,000 multi-brand outlets and 6,800+ points of sales in department stores across India. It is one of the country's largest fast fashion store brands. The Company also holds exclusive online and offline rights to the India network of California-based fast fashion brands.
Testing Implementation
With thousands of product variations in multiple categories, the client is not only a marketplace for its own brand but also for many others. Traditional automated approaches make testing such a category-heavy brand extremely time-consuming. The client owns several highly interactive app that helps their customers identify and make product selection based on their liking so that they may purchase the most suitable outfits. To do this, the client has implemented an incredibly helpful feature to help customers find the best fit. Testing this distinctive feature is no easy undertaking, as there are many questions that need to be answered and many possible permutations for the same. Test automation is rather complicated to derive and verify if the customer has received appropriate suggestions. The client's virtual assistant is one of the many discrete features available on the application. The chat assistant provides instantaneous support for any problems a consumer may be experiencing. Due to the wide variety of possible QnA interactions, testing this functionality was exhausting. The many promotions run by the client make its app quite dynamic. In addition, the client app features certain Games and Rewards where users may participate in competitions and earn coupons. This part too is adaptable and changes depending on the person using it. It's not easy to test such complex features. Finally, keeping the entire client app tested would require a large number of manual and automation test scripts to be maintained.Managing the defect lifecycle and implementing continuous integration and continuous deployment (CI/CD) for automation test scripts is also a significant challenge in eCommerce solutions.
FireFlink Solution
To begin, our team proactively uploaded 500 manual test cases to the FireFlink platform in just two days before embarking on script development for automation testing. The team's scope included developing scripts for both web and mobile (android & iOS) application test scripts. The FireFlink Finder made it easy to keep track of the App's numerous user interface components despite their dispersed placement across multiple tabs. The UI elements were organized into tabs for easier navigation. There was no need to re-record the locator for UI elements that appeared in several tabs; they could simply be shared between pages or screens. FireFlink Finder provided several possible locators for the UI elements and then, at runtime, prioritized the best one. The platform leverages AI-based natural language processing functionalities. When the NLP functions are invoked, the required activities are carried out by the backend code. The platform provides NLP suggestions based on the action/verification we provide. We were able to quickly and easily automate scripts for both the Web and Mobile using pre-configured natural language processing using simple English. We launched a test data importer that supports Excel, JSON, and XML so that we could test the same feature with a variety of input values that supported us in successfully test-automating virtual assistant scenarios. This data may be provided to the test cases that will be performed in iterations, with one set of data from each of these files being utilized for each iteration. Since the data was being fetched automatically, we could run test scripts with different inputs rapidly and with little human intervention. FireFlink is adaptive to UI dynamicity. So much so, even the game test automation is carried out using built-in executables. FireFlink managed coupon code validations by dynamically changing and updating the variable values all facilitated from a single script or an automation channel. Overall, adopting FireFlink made managing test cases easier because we didn't have to leave the platform to meet prerequisites for third-party tools. In addition, FireFlink allowed for the connection of the manual and its related automation test script, resulting in a solid tree structure for the test cases.