Ved Antani, Vice President of Engineering at Myntra, explains how the company uses deep personalization, a rich data platform, and data science to solve fashion problems in an exclusive interview with Express Computer. He also discloses Myntra's critical data integration measures.
- Give us a quick overview of Myntra's data and analytics journey.
How we create highly customized products for customers sets Myntra apart from other players in the Indian fashion e-commerce market. Our application's user experience is designed for the fashion industry, requiring new perspectives on personalization, data science, and user experience.
With this end in mind, Myntra's data science platforms and data engineering insights have developed into the fashion sector's first kind.
At Myntra, data is in our very DNA. Data is everything to us as a component of the leading e-commerce funnel. The story the data provides us significantly influences our decision-making for company or product definition. Our in-house AB testing and user segmentation solutions allow us to quickly experiment and evaluate product definition to create the greatest user experience. (To know more about how AB testing works, refer to the data science course in Chennai.)
The same data is used to improve our data science as well.
- Could you please list some of Myntra's data analytics use cases?
Growth Hacking is a brand-new feature that we just introduced. We quickly test out small- to medium-sized product concepts on various topics as part of Myntra's engineering department. In order to make the user journey more meaningful, these trials might, for example, offer a click-checkout experience or alter the post-order experience. Growth hacking is a technique to achieve specific business objectives or increase conversion and retention rates.
We do A-B testing for new features before releasing them to a fixed group of users to release Growth Hacking experiments. This is to guarantee that we track the performance of this feature, too. Myntra regularly publishes experiments and evaluates their results. They are also quick-moving experiments that we may distribute to users.
- What is the organization's analytics roadmap that you are creating? Have you thought of a specific plan of action?
The entire purchase and push order process would feature data science and machine learning, according to Myntra. The foundational analytics systems will gradually converge with the ML and data science platforms. As a result of our conviction that real-time ML models provide value to data science, Myntra is making significant investments in the development of ML platforms and real-time data science pipelines. As a result, Myntra is concentrating on developing the ML platform for data science model execution and deployment training. Also, we are creating a real-time portal to access real-time model feedback.
We use incredibly detailed data and data science to solve complex problems connected to fashion e-commerce at the scale that Myntra operates. To provide a wonderful customer experience, we gradually invest in these systems and add larger capabilities.
- What kind of infrastructure was designed specifically for analytics? How have you ensured that your data is accurate?
Before data is ingested into the warehouse, cleaned up, and processed, Myntra has numerous checks and balances in place. In order to make sure that even a little departure from the intended output is noted, these phases have telemetry, failure reports, and test risk coverage that are all clearly written out. We have gone through several iterations of this procedure because we are always enhancing the framework for monitoring and editing. We have a reliable alerting and testing structure to catch even the smallest shift or failure in our data pipeline.
I hope you got some insight into how myntra utilizes data science and analytics. If data science is something that interests you, have a look at an online Data science training in pune and acquire the knowledge of the latest tools.
