Nykaa Builds Unified Data Platform for Insights Generation Using AWS

Over the past decade, as Nykaa evolved from an exclusively beauty and personal care e-commerce platform to encompass various verticals, including grooming for men, fashion, eB2B and expanding its footprint internationally, data challenges arose. Each vertical operated with distinctly separate databases, which prevented the company from accessing a single source of sales, customer, and product inventory data.

Nykaa's siloed systems impeded the company’s ability to perform effective data analytics and generate business insights on its customers. Additionally, Nykaa’s multiple, disparate databases resulted in time-consuming management and increased the risk of contradictory data. The company also found that duplicating the same data across various locations increased its overall storage footprint and led to inefficient resource utilization.

Furthermore, for performance monitoring, Nykaa regularly had to collate and analyze clickstream and transactional data across various verticals. This resulted in data fragmentation across different databases. The company needed a team of 2-4 staff up to 3-4 weeks to generate the relevant business reports. Due to the time-consuming process, Nykaa would sometimes miss out on opportunities to reach specific customer groups at specific times.

In 2022, Nykaa embarked on its “One Nykaa, One Data” initiative to create a unified data ingestion, governance, and query framework. The company turned to AWS to streamline its operations into a single, scalable, reliable, and cost-effective data lake. The goal was to consolidate data from diverse sources, to enhance transparency for all stakeholders, yield superior business insights, and generate cross-selling and upselling opportunities.

In September 2022, Nykaa built a proof of concept (POC) by comparing the cost efficiency of Amazon EMR—a cloud-based big data solution for processing, analytics, and machine learning using open-source frameworks—against other cloud-based analytics solutions. After 3-4 weeks, Nykaa observed a 30 percent reduction in its core software expenses due to Amazon EMR’s elastic scaling capabilities and pay-as-you-go pricing model.

In Mar 2023, Nykaa began using Amazon EMR, Amazon Simple Storage Solution (Amazon S3), an object storage service, and cloud data warehousing service Amazon RedShift to build and deploy the ingestion layer of the data lake. As a cornerstone of its “One Nykaa, One Data” initiative, the data lake can now consolidate over 400 terabytes of unarchived data from various sources into a single data lake.

With autoscaling built into Amazon EMR, the company can now automatically scale its data processing workloads, facilitating up to 10,000 data ingestion jobs. Consequently, Nykaa’s business analytics team requires only 5 hours to generate a new business report from what took up to 2 weeks. Today, the team has scaled the data processing platform to generate over 1,500 sets of customer data. It also collects data on an hourly basis instead of daily, allowing Nykaa to swiftly respond to market trends and emerging customer preferences.

Nykaa uses AWS Lake Formation to build, manage, and secure the new data lake at the governance layer. With a secure way to gain deeper insights into customer behavior and preferences, Nykaa can build personalized marketing strategies and tailor product recommendations for different customer segments.

Finally, Nykaa uses Amazon Redshift and Amazon Athena, a serverless analytics service, to analyze petabyte-scale data at the query layer. Since both services are fully managed, Nykaa’s developers do not have to spend time operating and maintaining the data warehousing and query services.