Dollar Shave Club Uses AWS to Accelerate Data Analysis, Improve User Experience

Dollar Shave Club, an ecommerce company based in Venice, California, has delivered razors and other personal grooming products to millions of customers worldwide since 2011. 

The company runs its entire ecommerce platform on Amazon Web Services (AWS). “We were born on the AWS Cloud,” says Saritha Ivaturi, director of data systems at Dollar Shave Club. As it grew, the company increasingly sought ways to gain more in-depth knowledge of customer trends and products so it could provide a more personalized customer experience. To support this vision, Dollar Shave Club began using two Amazon Redshift clusters as its primary data warehouse. “Our data science and marketing teams use Amazon Redshift for ad hoc data analysis, looking at customer data to determine how to personalize products based on preferences and sales history,” Ivaturi says. 

However, as the volume of data grew, the company sought to optimize its analytics environment. “Our original cluster was 12 nodes. We wanted to find the best way to optimize both storage and compute using AWS services,” says Ivaturi. “We decided that meant moving from a large, traditional data warehouse to a more dynamic compute environment.”

To optimize its analytics solution, Dollar Shave Club adopted Amazon Redshift with a lake house architecture, integrating an 8-node data lake or raw data repository, based on Amazon Simple Storage Service (Amazon S3).

The company houses its low-latency data in Amazon Redshift and its moderate-latency data in Amazon S3, using the Amazon Redshift Spectrum feature to query over 60 terabytes of customer and product data directly from Amazon S3 daily, with no data loading. The company also mines the data for machine learning algorithms. “That means we can scale the entire environment more easily and we also reduce the large compute costs we had before,” says Ivaturi.

Dollar Shave Club stores all of its analytical data in the new data analytics solution. The company merged this data with third-party marketing data and created schemas in the AWS Glue Data Catalog, which allows employees to use business intelligence (BI) tools such as Tableau to get a consolidated data view for building reports. Users can also take advantage of the Amazon Redshift lake house because they can access data with their own BI tools, regardless of where the data is stored. Ivaturi says, “We have a lot of flexibility in that we can source external data securely and we can merge and curate that data so it’s ready for analytics and machine learning.”

Dollar Shave Club has accelerated its analytical capabilities by implementing its new environment. “Previously, it took us at least 8 hours to provide data insights to marketing and other teams. We have reduced reporting time to 5 minutes by democratizing our data using AWS technologies,” says Ivaturi. “All business users at our company can access fresh data with no restrictions and query that data using BI tools. Now, instead of producing three or four reports a week, a marketing employee can produce multiple reports a day.”

With faster analysis and reporting, Dollar Shave Club can make faster business decisions. For example, the analytics solution automatically sends notifications to the company’s email marketing platform. “Our marketing teams can make dynamic decisions on managing marketing campaigns or segmentation strategies,” says Ivaturi. The company can also move faster at the corporate level. “We experiment very quickly, and the AWS solution helps us be nimbler. We can quickly drill through billions of data records and see different combinations. Each week, for example, our executives look at the data from the solution to determine if we should pivot to a new strategy.”

By adopting a lake house approach based on Amazon S3 and Amazon Redshift, Dollar Shave Club has optimized its costs by separating compute from storage and reducing its data analytics cluster from 12 to 8 nodes. “We are saving about $300,000 every year by using Amazon S3 with Amazon Redshift,” says Ivaturi. “This is because we compressed the overall size of the cluster to use only the compute we need. We will shrink the cluster even more and save more money, all from switching from a traditional data warehouse and data lake approach to a lake house and dynamic compute environment.”

Dollar Shave Club is putting its savings back into research and new technology development. “We are investing the money in exploring and learning new AWS technologies such as AWS Lambda and AWS Database Migration Service, and we are taking advantage of new AWS services for machine learning and analytics such as Amazon Athena and AWS Glue Data Catalog,” says Ivaturi. “We are a data-driven company, and we can now put more resources into mining that data and finding new insights from it.”

Additionally, the organization can more easily scale its lake house environment. “This solution was designed with scalability in mind, so we can put more processes on top of Amazon S3 without seeing any degradation in performance,” says Sergey Zavgorodni, senior data architect at Dollar Shave Club.

Using insights from its data, Dollar Shave Club is offering a better customer experience. “We are taking the data we access and analyze through the Amazon Redshift lake house and enabling personalized product recommendations, website optimization, and new features,” says Ivaturi. “For instance, we’ve added new shops to the ecommerce site, so customers have more products to choose from now. We expect to gain new business insights as we add more data into the environment, which means we can further differentiate our business by giving our customers more.”

To learn more, visit aws.amazon.com/redshift.