Benefit Cosmetics Case Study

Customer Profile

Benefit Cosmetics (a subsidiary of Louis Vuitton) is a prominent cosmetic brand based in San Francisco, CA. They sell at over 3,000 counters in more than 40 countries. Benefit Cosmetics vision is to continue to grow their footprint globally. A key marketing channel is digital that includes a highly engaging web platform for learning about its products and services. As an industry leader, Benefit Cosmetics emphasizes the importance of giving back and is involved in many charities across the globe.

The Opportunity

Benefit Cosmetics has made business agility and engineering velocity strategic objectives. To enable rapid, parallel development, the company sought to refactor its conventionally-hosted site infrastructure around core AWS platform technologies. This included both the content management systems for the company as well as there production e-commerce website. The company was committed to a cloud-first implementation and wisely invested in extensive automation from the beginning of the engagement. The company engaged Cascadeo, an AWS Premier Consulting Partner, to design and implement an AWS-native solution.

The Partner

Cascadeo is a cloud DevOps engineering firm focused on mid-market to enterprise customers that require high levels of availability, performance, security, and scalability. We specialize in environments that require multi-technology dependencies, hybrid cloud solutions and connectivity, complex operational requirements, infrastructure automation, self-healing capabilities, predictive analytics and continuous deployments.

The Solution

Cascadeo and Benefit agreed to a DevOps/Agile-driven, AWS-native solution architecture. Because of the criticality of the system, and the company's commitment to innovation in online sales, the system was designed for Agile-style iterative development and frequent redeployment to multiple environments. This meant fully repeatable deployments and end-to-end configuration management, implemented using Jenkins, CloudFormation, and Chef. Containerized application workloads run on Kubernetes, and operational telemetry is managed with Grafana and InfluxDB. Benefit Cosmetics developers can deploy, and redeploy, the entire system at-will. Code check-ins automatically trigger deployments, and live demos of new features can be shared internally or promoted to the production environment within minutes of their implementation.

AWS Services Used

AWS RDS with AuroraDB, AWS ElastiCache with Redis, AWS EFS for shared storage, AWS CloudFormation, AWS S3, AWS ECR, AWS Route53, AWS ELB, AWS EC2, AWS Lambda, AWS Simple AD, AWS CloudWatch

How We Won

Cascadeo initially was engaged in improving existing Chef cookbooks and recipes around best current practices. During the initial engagement, Cascadeo engineers proposed a containerized / microservices approach, and the relationship expanded to include application-layer refactoring around AWS platform technologies. By taking a strategic, business-value-first approach to the project, Cascadeo has dramatically improved Benefit Cosmetics’ time to market and business agility.

Benefit to Client

  • Engineering velocity, time to market, and business agility are all dramatically improved through frequent, full-stack deployment automation. Benefit Cosmetics can take an idea from inception to live demo to production in days, rather than weeks or months.
  • As additional developers join the growing team, each adopts the same development environment and DevOps-style deployment automation, eliminating the risk of subtle differences in environments causing downstream problems as code is promoted to production.
  • We reduced the instance deployment from hours to minutes for client’s development teams.
  • By implementing blue/green deployment strategy, we dramatically reduced errors in the production deployment cycle.
  • All code release are aligned with container image version for easy management
  • Production deployment is done with one-click approach using Jenkins and Development instance are integrated with Slack with a simple slash command.