CRM software company leverages Delivery Automation to release faster

Introduction

This case-study explores the journey of a CRM software company based in Bengaluru, India, who provides customer relationship and contact center solutions to some of the largest businesses globally.

They configure their solutions uniquely for each client, with each client having their dedicated installation within their private infrastructure.

Problem

1. Manual on-site deployments causing delays and incurring additional expenses due to resource costs.
2. Slow release and upgrade cycles impacting the company's ability to release new features promptly.
3. Dependency on manual processes and people, leading to errors and inconsistencies in deployments.

Objectives

1. Implement Infrastructure as Code (IaC) for remote deployments with an automated Delivery Management System.
2. Achieve faster release and upgrade cycles through automated CI/CD pipelines to build and release packages.
3. Introduce pull-based deployments, which will allow client instances to download packages and update themselves whenever there is a software update.

Implementation

1. Delivery Management System:
   • Developed a Delivery Management System (DMS portal) to manage client instances and their releases.
   • Remotely deploy and configure services on a client's infrastructure without being present in the client’s location.
   • Backup, update, and roll back releases manually, as and when necessary,
2. IaC Implementation:
   • Adopted Terraform for defining and configuring infrastructure for client instances.
   • Established version-controlled IaC templates for client environments, managed in GitLab repositories.
3. Automated CI/CD Pipelines:
   • Integrated GitLab CI/CD pipelines for automated build, testing and releasing packages to a central package registry.
   • Implemented automated testing suites to ensure the reliability of application updates.
4. Dependency Reduction:
   • Introduced monitoring and alerting tools for proactive issue identification.
   • Automated rollback mechanisms in case of release failures to minimize downtime.
   • Conducted comprehensive training for the operations team to handle the new automated processes.

Results

1. Remote Deployment Metrics:
   • Achieved 90% reduction in deployment time through remote IaC deployments.
   • On-site deployment costs reduced by 95%.
2. Release and Upgrade Cycle Metrics:
   • Reduced release cycles from quarterly to bi-weekly.
   • 95% success rate in automated deployments, minimizing errors and inconsistencies.
3. Dependency Reduction Metrics:
   • 78% reduction in manual interventions during deployment.
   • Proactive identification and resolution of potential issues led to an 85% decrease in post-deployment incidents.

Challenges

1. Team Adaptation:
   • Overcoming resistance to change and ensuring team members are leveraging in IaC and automated processes.
   • Learning: Continuous training and support are crucial for successful adoption.
2. Integration Complexity:
   • Ensuring seamless integration between AWS services, GitLab, and internal processes.
   • Learning: Rigorous testing and collaboration between development and operations teams are essential.
3. Rollback Mechanism Fine-Tuning:
   • Adjusting automated rollback mechanisms to handle various deployment scenarios.
   • Learning: Regular testing and refinement of rollback procedures are necessary for minimizing downtime.

Conclusion

The successful adoption of automated Delivery Management has streamlined deployment processes, significantly reduced costs, and improved release and upgrade cycles. The company now operates with increased agility, delivering timely updates to customers while maintaining high reliability.