Accelerating Feature delivery by decoupling deployments and feature releases

Overview

In 2024, a global insurance organisation partnered with DevOps1 to accelerate their feature release process. The organisation sought to provide application and delivery teams with the capability for decoupled and parallel feature releases, enabling them to deliver value to customers more frequently and efficiently.

Challenges

At present, the organisation lacks a strategic Feature Toggle, Release and Environment Platform to accelerate and uplift development and delivery process across platforms and delivery teams. This engagement assessed the current state and provide recommendation on strategic solution to address current challenges or gaps with feature delivery and Release, Environment capability.

During the assessment, DevOps1 identified that the organisation architecture and technology stack presented significant challenges to the feature release process. The organisation relied heavily on SaaS applications and different technology stacks, which made it difficult to ensure consistency across feature releases.

Solutions

Our approach to addressing these challenges involved a detailed analysis of the organisation's technology stack and architecture, identifying areas for standardisation and consolidation to improve consistency and efficiency in the feature release process.

A two-phase approach was adopted to achieve this goal. The first phase focused on the assessment of the current state of the organisation's feature release process, identifying bottlenecks and areas for improvement. The second phase involved the design and implementation of a feature release acceleration plan, which included a set of initiatives to streamline the feature release process and improve the efficiency of the organisation's delivery teams.

Outcomes

• Increased release frequency to at least fortnightly and increased revenue, decreased costs
• Reduction of environments will impact reduced cost, time to market
• Improve quality and reliability and avoid down-time reducing engineering effort required per feature rollout
• Ability to automatically measure the impact of engineering work providing faster problem detection to protect the organisation
• Reduce the risk of code failure and effects of broken deployments with simple rollbacks

Deliver

• 34 Discovery sessions with multiple stakeholders
• Analysis and Review of multiple tools with Proof of concepts to validate feasibility
• Procurement of LaunchDarkly licenses
• Implementation and Blueprint of LaunchDarkly for SaaS systems and inhouse development applications