Improved the Retention Rate by 70% by delivering Personalized user recommendations for an American mass media house conglomerate
Viewership details of each user are captured from various partners such as Hulu, Youtube, AppleTV, etc. and stored in Redshift
Highlights
70%
Improved Customer Retention Rate
3–5x
Improvement in BI Reporting load times
40%
Project delivery time saved
20%
Reduced TCO up to
Client Profile
The client is an American mass media and entertainment conglomerate. It is one of the world’s leading media and entertainment companies in the development, production, and marketing of entertainment, news, and information to a global audience.
Business challenges
- Lack of Personalized user experience due to the absence of Personalized recommendation Engine
- Need to Fetch information from different brands / partners (Youtube, Hulu, etc.) for a single user and provide recommendations which maps to the right brand
- Revenue impact due to high churn rate – the absence of personalized content
- Lack of business insights for deeper perspective and understanding
Agilisium Solution
- Viewership details of each user are captured from various partners such as Hulu, Youtube, AppleTV, etc. and stored in Redshift
- The Recommendation Engine arrives at a ranked order of likely shows to be viewed by the user
- The top 10 recommended shows by the user is stored in DynamoDB
- When the users log in, the app uses the recommendations to create personalized offers
Tech-stacks Used
AWS – S3
Redshift
Business outcome
Recommendation Engine
- Viewership details are stored in Redshift
- The top 10 recommended shows are stored in DynamoDB, and the same is retrieved to create personalized offers while user login
- S3 Storage is used while ML algorithms are running and get stored in S3
Deployment Automation
- Applications are deployed to lambda serverless environments and hosted on containers in AWS ECS
- AWS Auto Scaling is enabled to scale in and scale out ECS clusters and services
- AWS Cloudformation and Terraform are being used for Infrastructure Automation
- Deployment is done using CircleCI with automated pipelines
Monitoring & Logging
- AWS XRay is used for all serverless lambda applications, and it helps in monitoring most of the applications
- For applications running on containers, AWS CloudWatch metrics are being leveraged
- Can Integrate with third-party monitoring tools like SignalFx, NewRelic, and Splunk
- Application-level health checks are also enabled and are being monitored using Grafana
Security
- IAM best practices and principles are followed
- AWS Certificate Manager is being used for the load balancer and CDN certificates
- All Data stores are in private subnet
- Amazon KMS is used for encryption of data at rest