API development and system configuration for integrating real-time data with various systems/OS in Europe.
Before opening the platform in March 2019, GUHADA worked on developing API integraion technology for three years. It was based on the judgment that if a system was created that allows domestic customers to check stocks in European boutiques in real time and place orders immediately, they would be freed from the controversy over counterfeit products and would be able to solve the chronic “shortage” problem of luxury commerce. Here, the shortage problem refers to cases in which a customer is unilaterally notified of out-of-stock or cancellation due to a lack of inventory after ordering, and is a major cause of customer churn.
First, the modeling work was preceded by structuring big data provided by boutiques using different systems and operating systems into the patterns and architecture of GUHADA. Building a data pipeline that automates the process of extracting, verifying, and loading product data was essential due to the nature of the luxury business, which handles more than 200,000 items from over 1,500 brands and a vast variety of items. From real-time inventory information, which is the core of service operation, to product names, categories, detailed options, images and components, the format and structure of various information was documented and mapped so that they could flow logically.
In particular, the most sensitive data, commodity prices, has been algorithmized to minimize the risk of fluctuations that may occur due to various variables such as exchange rates, freight rates, discounts, and tariffs. Then, using the self-developed big data analysis engine (Watcher Algorithm), based on the same product data collected from all over the world, optimal pricing that secures market competitiveness for all product data by dividing the cost-to-margin section and competitiveness section has been applied. And throughout the above process, high-performance big data processing technology was used. It was a task to ensure the accuracy of the data by updating more than 30,000 local data every two hours and matching it with the operating system of GUHADA in a short time. At this time, data cleansing was also performed to automatically remove abnormal data or redundant data.
The big data collected through the above process is stored in a cloud cache server and database. Of course, the cloud is configured in various forms and systems to prevent data silos, and since data scaling technology tailored to product data traffic was developed in the initial system planning stage, potential risks such as data loss, inconsistency, and errors could be minimized.
Finally, when a product order is received from a customer, GUHADA have developed and are using its own admin system to automatically place an order through boutique API access, and share data dashboards with local partner boutiques to create big data on brand, product search information, price, color, and size. It was implemented to enable two-way communication to check and analyze order information. Through this, both companies can predict customer demand for pre-orders or stock orders for the next season and fashion trends based on data.
Thank you for your support!
TEMCO team.
